types.py

import sys
from datetime import datetime
from enum import Enum, EnumMeta
from typing import (
    TYPE_CHECKING,
    Annotated,
    Any,
    Dict,
    List,
    Literal,
    Optional,
    Sequence,
    Tuple,
    Union,
)
from urllib.parse import parse_qs, urlparse
from warnings import warn

if TYPE_CHECKING:
    from .transcriber import Transcript

try:
    # pydantic v2 import
    from pydantic import BaseModel, ConfigDict, Field, field_validator
    from pydantic_settings import BaseSettings, SettingsConfigDict

    pydantic_v2 = True
except ImportError:
    # Python 3.14+ requires Pydantic V2
    if sys.version_info >= (3, 14):
        raise ImportError(
            "Python 3.14 or greater requires Pydantic V2 and pydantic-settings. "
            "Please install with: pip install 'pydantic>=2.0' 'pydantic-settings>=2.0'"
        ) from None

    # pydantic v1 import (fallback for Python < 3.14)
    from pydantic.v1 import BaseModel, BaseSettings, ConfigDict, Field, validator

    pydantic_v2 = False

from typing_extensions import Self


class AssemblyAIError(Exception):
    """
    Base exception for all AssemblyAI errors
    """

    def __init__(self, message: str, status_code: Optional[int] = None):
        super().__init__(message)
        self.status_code = status_code


class TranscriptError(AssemblyAIError):
    """
    Error class when a transcription fails
    """


class RedactedAudioIncompleteError(AssemblyAIError):
    """
    Error class when a PII-redacted audio URL is requested
    before the file has finished processing
    """


class RedactedAudioExpiredError(AssemblyAIError):
    """
    Error class when a PII-redacted audio URL is requested
    but the file has expired and is no longer available
    """


class RedactedAudioUnavailableError(AssemblyAIError):
    """
    Error class when a PII-redacted audio file is requested
    but it is not available at the given URL
    """


class LemurError(AssemblyAIError):
    """
    Error class when a Lemur request fails
    """


class SyncTranscriptError(AssemblyAIError):
    """
    Error raised when a synchronous transcription request fails.

    Carries a machine-readable `error_code` — the snake_cased problem-details
    `title` from the server (e.g. `bad_audio`, `audio_too_large`,
    `capacity_exceeded`, `inference_timeout`) — when present, and
    `retry_after` (seconds) for 429/503 responses that include a
    `Retry-After` header.
    """

    def __init__(
        self,
        message: str,
        status_code: Optional[int] = None,
        error_code: Optional[str] = None,
        retry_after: Optional[int] = None,
    ):
        super().__init__(message, status_code)
        self.error_code = error_code
        self.retry_after = retry_after


class Sourcable:
    """
    A base class for all sourcable objects

    Currently, only `Transcript` is sourcable
    """


class Settings(BaseSettings):
    """
    Settings for the AssemblyAI client
    """

    api_key: Optional[str] = None
    "The API key to authenticate with"

    http_timeout: float = 30.0
    "The default HTTP timeout for general requests"

    keepalive_expiry: Optional[float] = None
    "Seconds an idle connection is kept in the HTTP pool before being closed. None uses httpx's default (5s). Raise it (e.g. to 120) so a connection opened by `SyncTranscriber.warm()` stays reusable across an in-progress recording."

    base_url: str = "https://api.assemblyai.com"
    "The base URL for the AssemblyAI API"

    sync_base_url: str = "https://sync.assemblyai.com"
    "The base URL for the synchronous transcription API (used by `SyncTranscriber`)"

    sync_http_timeout: float = 60.0
    "The HTTP timeout for synchronous transcription requests. Kept above the server's 30s deadline so the client doesn't race it."

    polling_interval: float = Field(default=3.0, gt=0.0)
    "The default polling interval for long-running requests (e.g. polling the `Transcript`'s status)"

    if pydantic_v2:
        model_config = SettingsConfigDict(env_prefix="assemblyai_")
    else:

        class Config:
            env_prefix = "assemblyai_"


class TranscriptStatus(str, Enum):
    """
    Transcript status
    """

    queued = "queued"
    processing = "processing"
    completed = "completed"
    error = "error"


class DeprecatedLanguageCodeMeta(EnumMeta):
    def __getattribute__(self, item):
        # Deprecate all 20 possible values
        languages = [
            "de",
            "en",
            "en_au",
            "en_uk",
            "en_us",
            "es",
            "fi",
            "fr",
            "hi",
            "it",
            "ja",
            "ko",
            "nl",
            "pl",
            "pt",
            "ru",
            "tr",
            "uk",
            "vi",
            "zh",
        ]
        if item in languages:
            warn(
                "LanguageCode Enum is deprecated and will be removed in 1.0.0. Use a string instead.",
                DeprecationWarning,
                stacklevel=2,
            )

        return EnumMeta.__getattribute__(self, item)


class LanguageCode(str, Enum, metaclass=DeprecatedLanguageCodeMeta):
    """
    DeprecationWarning: LanguageCode is deprecated and will be removed in 1.0.0. Use a string instead.

    Supported languages for transcribing audio.
    """

    de = "de"
    "German"

    de_ch = "de_ch"
    "Swiss German"

    en = "en"
    "Global English"

    en_au = "en_au"
    "Australian English"

    en_uk = "en_uk"
    "British English"

    en_us = "en_us"
    "English (US)"

    es = "es"
    "Spanish"

    fi = "fi"
    "Finnish"

    fr = "fr"
    "French"

    hi = "hi"
    "Hindi"

    it = "it"
    "Italian"

    ja = "ja"
    "Japanese"

    ko = "ko"
    "Korean"

    nl = "nl"
    "Dutch"

    pl = "pl"
    "Polish"

    pt = "pt"
    "Portuguese"

    ru = "ru"
    "Russian"

    tr = "tr"
    "Turkish"

    uk = "uk"
    "Ukrainian"

    vi = "vi"
    "Vietnamese"

    zh = "zh"
    "Chinese"


class WordBoost(str, Enum):
    low = "low"
    default = "default"
    high = "high"


class PIIRedactedAudioQuality(str, Enum):
    mp3 = "mp3"
    wav = "wav"


class PIIRedactedAudioMethod(str, Enum):
    silence = "silence"


class RedactPiiAudioOptions(BaseModel):
    """Options for controlling PII audio redaction behavior."""

    override_audio_redaction_method: Optional[PIIRedactedAudioMethod] = Field(
        None,
        description="Override the default audio redaction method (beep) with an alternative (silence).",
    )


class EntityType(str, Enum):
    """
    Used for AssemblyAI's Entity Detection feature.

    See: https://www.assemblyai.com/docs/audio-intelligence/entity-detection
    """

    account_number = "account_number"
    "Customer account or membership identification number (e.g., Policy No. 10042992; Member ID: HZ-5235-001)"

    banking_information = "banking_information"
    "Banking information, including account and routing numbers (e.g., Acct. No.: 012345-67)"

    blood_type = "blood_type"
    "Blood type (e.g., O-, AB positive)"

    credit_card_cvv = "credit_card_cvv"
    "Credit card verification code (e.g., CVV: 080)"

    credit_card_expiration = "credit_card_expiration"
    "Expiration date of a credit card (e.g., Expires: July 2023; Exp: 02/28)"

    credit_card_number = "credit_card_number"
    "Credit card number (e.g., 0123 0123 0123 0123)"

    date = "date"
    "Specific calendar date (e.g., December 18)"

    date_interval = "date_interval"
    "Broader time periods, including date ranges, months, seasons, years, and decades (e.g., 2020-2021; 5-9 May; January 1984 )"

    date_of_birth = "date_of_birth"
    "Date of Birth (e.g., Date of Birth: March 7,1961)"

    drivers_license = "drivers_license"
    "Driver's license number (e.g., DL# 356933-540)"

    drug = "drug"
    "Medications, vitamins, or supplements (e.g., Advil, Acetaminophen, Panadol)"

    duration = "duration"
    "Periods of time, specified as a number and a unit of time (e.g., 8 months; 2 years)"

    email_address = "email_address"
    "Email address (e.g., support@assemblyai.com)"

    event = "event"
    "Name of an event or holiday (e.g., Olympics, Yom Kippur)"

    filename = "filename"
    "Names of computer files, including the extension or filepath (e.g., Taxes/2012/brad-tax-returns.pdf)"

    gender = "gender"
    "Terms indicating gender identity (e.g., female, male, non-binary)"

    gender_sexuality = "gender_sexuality"
    "Terms indicating gender identity or sexual orientation, including slang terms (e.g., female; bisexual; trans)"

    healthcare_number = "healthcare_number"
    "Healthcare numbers and health plan beneficiary numbers (e.g., Policy No.: 5584-486-674-YM)"

    injury = "injury"
    "Bodily injury (e.g., I broke my arm, I have a sprained wrist)"

    ip_address = "ip_address"
    "Internet IP address, including IPv4 and IPv6 formats (e.g., 192.168.0.1)"

    language = "language"
    "Name of a natural language (e.g., Spanish, French)"

    location = "location"
    "Any Location reference including mailing address, postal code, city, state, province, country, or coordinates (e.g., Lake Victoria, 145 Windsor St., 90210)"

    location_address = "location_address"
    "Mailing address (e.g., 123 Main Street, Apartment 4B)"

    location_address_street = "location_address_street"
    "Street address (e.g., 123 Main Street)"

    location_city = "location_city"
    "City name (e.g., San Francisco, New York)"

    location_coordinate = "location_coordinate"
    "Geographic coordinates (e.g., 37.7749° N, 122.4194° W)"

    location_country = "location_country"
    "Country name (e.g., United States, Canada)"

    location_state = "location_state"
    "State or province name (e.g., California, Ontario)"

    location_zip = "location_zip"
    "Postal or ZIP code (e.g., 94102, M5V 3A8)"

    marital_status = "marital_status"
    "Terms indicating marital status (e.g., Single, common-law, ex-wife, married)"

    medical_condition = "medical_condition"
    "Name of a medical condition, disease, syndrome, deficit, or disorder (e.g., chronic fatigue syndrome, arrhythmia, depression)"

    medical_process = "medical_process"
    "Medical process, including treatments, procedures, and tests (e.g., heart surgery, CT scan)"

    money_amount = "money_amount"
    "Name and/or amount of currency (e.g., 15 pesos, $94.50)"

    nationality = "nationality"
    "Terms indicating nationality, ethnicity, or race (e.g., American, Asian, Caucasian)"

    number_sequence = "number_sequence"
    "Numerical PII (including alphanumeric strings) that doesn't fall under other categories"

    occupation = "occupation"
    "Job title or profession (e.g., professor, actors, engineer, CPA)"

    organization = "organization"
    "Name of an organization (e.g., CNN, McDonalds, University of Alaska, Northwest General Hospital)"

    organization_medical_facility = "organization_medical_facility"

    passport_number = "passport_number"
    "Passport numbers, issued by any country (e.g., PA4568332; NU3C6L86S12)"

    password = "password"
    "Account passwords, PINs, access keys, or verification answers (e.g., 27%alfalfa, temp1234, My mother's maiden name is Smith)"

    person_age = "person_age"
    "Number associated with an age (e.g., 27, 75)"

    person_name = "person_name"
    "Name of a person (e.g., Bob, Doug Jones, Dr. Kay Martinez, MD)"

    phone_number = "phone_number"
    "Telephone or fax number (e.g., +4917643476050)"

    physical_attribute = "physical_attribute"
    "Distinctive bodily attributes, including terms indicating race (e.g., I'm 190cm tall, He has black hair)"

    political_affiliation = "political_affiliation"
    "Terms referring to a political party, movement, or ideology (e.g., Republican, Liberal)"

    religion = "religion"
    "Terms indicating religious affiliation (e.g., Hindu, Catholic)"

    sexuality = "sexuality"
    "Terms indicating sexual orientation (e.g., heterosexual, gay, bisexual)"

    statistics = "statistics"
    "Medical statistics (e.g., 18%, 18 percent)"

    time = "time"
    "Expressions indicating clock times (e.g., 19:37:28, 10pm EST)"

    url = "url"
    "Internet addresses (e.g., www.assemblyai.com)"

    us_social_security_number = "us_social_security_number"
    "Social Security Number or equivalent (e.g., 078-05-1120, ***-***-3256)"

    username = "username"
    "Usernames, login names, or handles (e.g., @AssemblyAI)"

    vehicle_id = "vehicle_id"
    "Vehicle identification numbers (VINs), vehicle serial numbers, and license plate numbers (e.g., 5FNRL38918B111818, BIF7547)"

    zodiac_sign = "zodiac_sign"
    "Names of Zodiac signs (e.g., Aries, Taurus)"

    # BETA - only english
    corporate_action = "corporate_action"
    "Corporate actions (e.g., merger, acquisition, IPO)"

    day = "day"
    "Day reference (e.g., Monday, Friday)"

    effect = "effect"
    "Effect or result (e.g., increase, decrease)"

    financial_metric = "financial_metric"
    "Financial metrics (e.g., revenue, profit margin, EBITDA)"

    medical_code = "medical_code"
    "Medical codes (e.g., ICD-10, CPT codes)"

    month = "month"
    "Month reference (e.g., January, February)"

    organization_id = "organization_id"
    "Organization identification numbers (e.g., EIN, company registration number)"

    product = "product"
    "Product names (e.g., iPhone, Tesla Model 3)"

    project = "project"
    "Project names (e.g., Project Apollo, Manhattan Project)"

    trend = "trend"
    "Trend indicators (e.g., upward trend, downward trend)"

    year = "year"
    "Year reference (e.g., 2023, 1999)"


# EntityType and PIIRedactionPolicy share the same values
PIIRedactionPolicy = EntityType
"""
Used for AssemblyAI's PII Redaction feature.

See: https://www.assemblyai.com/docs/audio-intelligence/pii-redaction
"""


class PIISubstitutionPolicy(str, Enum):
    """
    Used for AssemblyAI's PII Redaction feature.

    See: https://www.assemblyai.com/docs/audio-intelligence/pii-redaction
    """

    hash = "hash"
    "PII that is detected is replaced with a hash - #. For example, I'm calling for John is replaced with ####. (Applied by default)"

    entity_name = "entity_name"
    "PII that is detected is replaced with the associated policy name. For example, John is replaced with [PERSON_NAME]. This is recommended for readability."


class SummarizationModel(str, Enum):
    """
    Used for AssemblyAI's Summarization feature.

    See: https://www.assemblyai.com/docs/audio-intelligence/summarization
    """

    informative = "informative"
    """
    Best for files with a single speaker such as presentations or lectures.

    Supported Summarization Types:
        - `bullets`
        - `bullets_verbose`
        - `headline`
        - `paragraph`

    Required Parameters:
        - `punctuate`: `True`
        - `format_text`: `True`
    """

    conversational = "conversational"
    """
    Best for any 2 person conversation such as customer/agent or interview/interviewee calls.

    Supported Summarization Types:
        - `bullets`
        - `bullets_verbose`
        - `headline`
        - `paragraph`

    Required Parameters:
        - `punctuate`: `True`
        - `format_text`: `True`
        - `speaker_labels` or `dual_channel` set to `True`
    """

    catchy = "catchy"
    """
    Best for creating video, podcast, or media titles.

    Supported Summarization Types:
        - `headline`
        - `gist`

    Required Parameters:
        - `punctuate`: `True`
        - `format_text`: `True`
    """


class SummarizationType(str, Enum):
    """
    Used for AssemblyAI's Summarization feature.

    See: https://www.assemblyai.com/docs/audio-intelligence/summarization
    """

    bullets = "bullets"
    "A bulleted summary with the most important points."

    bullets_verbose = "bullets_verbose"
    "A longer bullet point list summarizing the entire transcription text."

    gist = "gist"
    "A few words summarizing the entire transcription text."

    headline = "headline"
    "A single sentence summarizing the entire transcription text."

    paragraph = "paragraph"
    "A single paragraph summarizing the entire transcription text."


class SpeechModel(str, Enum):
    """
    Used for AssemblyAI's Speech Model feature.
    """

    best = "best"
    "The best model optimized for accuracy."

    nano = "nano"
    "A lightweight, lower cost model for a wide range of languages."

    slam_1 = "slam-1"
    "A Speech Language Model optimized explicitly for speech-to-text tasks"

    universal = "universal"
    "The model optimized for accuracy, low latency, ease of use, and multi-language support"


class LanguageDetectionOptions(BaseModel):
    """Options for controlling the behavior or Automatic Language Detection"""

    expected_languages: Optional[List[str]] = Field(
        None, description="A list of languages that the audio could be expected to be."
    )
    fallback_language: Optional[str] = Field(
        None,
        description="The language to fallback to in case the language detection does not predict any of the expected ones.",
    )
    code_switching: Optional[bool] = Field(
        None,
        description="Enable code switching detection for multilingual transcription.",
    )
    code_switching_confidence_threshold: Optional[float] = Field(
        None,
        description="The confidence threshold for code switching detection. Valid values are in the range [0,1] inclusive.",
    )
    on_low_language_confidence: Optional[str] = Field(
        None,
        description='Controls behavior when language confidence is below threshold. Either "error" (default) or "fallback".',
    )
    swiss_german: Optional[bool] = Field(
        None,
        description="If True, route German language requests as Swiss German audio. Only applies when language_detection is enabled and German is detected.",
    )
    language_detection_model: Optional[str] = Field(
        None,
        description="The language detection model to use.",
    )


class CodeSwitchingLanguage(BaseModel):
    """Code switching language detection result"""

    language_code: str
    "The language code detected"

    confidence: float
    "The confidence score for this language detection, between 0.0 and 1.0"


class LanguageDetectionResults(BaseModel):
    """Language detection results including code switching languages"""

    code_switching_languages: Optional[List[CodeSwitchingLanguage]] = None
    "List of detected languages with confidence scores when code switching is enabled"


class SpeakerType(str, Enum):
    """
    Speaker identification type for speech understanding
    """

    role = "role"
    "Identify speakers by their role"

    name = "name"
    "Identify speakers by their name"


class SpeakerIdentificationRequest(BaseModel):
    """
    Speaker identification configuration for speech understanding
    """

    speaker_type: SpeakerType
    "The type of speaker identification to perform"

    known_values: Optional[List[str]] = None
    "Known speaker values (required when speaker_type is 'role')"

    speakers: Optional[List[Dict[str, Any]]] = None
    "Known speaker definitions with optional descriptions for improved accuracy"


class TranslationRequest(BaseModel):
    """
    Translation configuration for speech understanding
    """

    target_languages: List[str]
    "List of target language codes to translate the transcript into"

    formal: Optional[bool] = False
    "Whether to use formal language in translations (default: False)"

    match_original_utterance: Optional[bool] = False
    "Whether to match the original utterance structure in translations (default: False)"


class CustomFormattingRequest(BaseModel):
    """
    Custom formatting configuration for speech understanding
    """

    date: Optional[str] = None
    "Custom date format pattern (e.g., 'mm/dd/yyyy')"

    phone_number: Optional[str] = None
    "Custom phone number format pattern (e.g., '(xxx)xxx-xxxx')"

    email: Optional[str] = None
    "Custom email format pattern (e.g., 'username@domain.com')"


class SpeechUnderstandingFeatureRequests(BaseModel):
    """
    Speech understanding feature requests
    """

    speaker_identification: Optional[SpeakerIdentificationRequest] = None
    "Speaker identification configuration"

    translation: Optional[TranslationRequest] = None
    "Translation configuration"

    custom_formatting: Optional[CustomFormattingRequest] = None
    "Custom formatting configuration"


class SpeechUnderstandingRequest(BaseModel):
    """
    Speech understanding request configuration for LLM Gateway features
    """

    request: Optional[SpeechUnderstandingFeatureRequests] = None
    "The speech understanding feature requests"


class SpeakerIdentificationResponse(BaseModel):
    """
    Speaker identification response containing status and mapping
    """

    status: str
    "Status of the speaker identification feature (e.g., 'success')"

    mapping: Optional[Dict[str, str]] = None
    "Mapping of original speaker labels to identified speaker labels"


class CustomFormattingResponse(BaseModel):
    """
    Custom formatting response containing mapping and formatted texts
    """

    mapping: Optional[Dict[str, str]] = None
    "Mapping of original entities to formatted entities"

    formatted_text: Optional[str] = None
    "Full transcript text with formatted entities"

    formatted_utterances: Optional[List[Dict[str, Any]]] = None
    "List of utterances with formatted text"

    status: str
    "Status of the custom formatting feature"


class KeytermsPromptOptions(BaseModel):
    """Keyterms prompt options for controlling keyterms boosting behavior"""

    keyterms_match_strength: Optional[Literal["high", "standard"]] = Field(
        None,
        description="Level of match strength for keyterms boosting ('high' or 'standard')",
    )


class TranslationResponse(BaseModel):
    """
    Translation response containing status
    """

    status: str
    "Status of the translation feature"


class SpeechUnderstandingFeatureResponses(BaseModel):
    """
    Speech understanding feature responses grouped together
    """

    speaker_identification: Optional[SpeakerIdentificationResponse] = None
    "Speaker identification results including status and mapping"

    translation: Optional[TranslationResponse] = None
    "Translation results"

    custom_formatting: Optional[CustomFormattingResponse] = None
    "Custom formatting results"


class SpeechUnderstandingResponse(BaseModel):
    """
    Speech understanding response containing both request and response
    """

    request: Optional[SpeechUnderstandingFeatureRequests] = None
    "The original speech understanding request"

    response: Optional[SpeechUnderstandingFeatureResponses] = None
    "The speech understanding feature responses"


class SpeakerOptions(BaseModel):
    """
    Speaker options for controlling speaker diarization parameters
    """

    min_speakers_expected: Optional[int] = Field(
        None, ge=1, description="Minimum number of speakers expected in the audio"
    )
    max_speakers_expected: Optional[int] = Field(
        None, ge=1, description="Maximum number of speakers expected in the audio"
    )
    use_two_stage_clustering: Optional[bool] = Field(
        None,
        description="Enable or disable two-stage clustering for speaker diarization",
    )
    long_file_diarization_method: Optional[Literal["standard", "experimental"]] = Field(
        None,
        description="Diarization method for long files. Options: standard (default), experimental",
    )

    if pydantic_v2:

        @field_validator("max_speakers_expected")
        @classmethod
        def validate_max_speakers(cls, v, info):
            if v is not None and info.data.get("min_speakers_expected") is not None:
                min_speakers = info.data["min_speakers_expected"]
                if v < min_speakers:
                    raise ValueError(
                        "max_speakers_expected must be greater than or equal to min_speakers_expected"
                    )
            return v

    else:

        @validator("max_speakers_expected")
        def validate_max_speakers(cls, v, values):
            if v is not None and values.get("min_speakers_expected") is not None:
                min_speakers = values["min_speakers_expected"]
                if v < min_speakers:
                    raise ValueError(
                        "max_speakers_expected must be greater than or equal to min_speakers_expected"
                    )
            return v


class RawTranscriptionConfig(BaseModel):
    language_code: Optional[Union[str, LanguageCode]] = None
    """
    The language of your audio file. Possible values are found in Supported Languages.

    The default value is "en_us".
    """

    punctuate: Optional[bool] = None
    "Enable Automatic Punctuation"

    format_text: Optional[bool] = None
    "Enable Text Formatting"

    dual_channel: Optional[bool] = None
    "Enable Dual Channel transcription"

    multichannel: Optional[bool] = None
    "Enable Multichannel transcription"

    webhook_url: Optional[str] = None
    "The URL we should send webhooks to when your transcript is complete."
    webhook_auth_header_name: Optional[str] = None
    "The name of the header that is sent when the `webhook_url` is being called."
    webhook_auth_header_value: Optional[str] = None
    "The value of the `webhook_auth_header_name` that is sent when the `webhook_url` is being called."

    audio_start_from: Optional[int] = None
    "The point in time, in milliseconds, to begin transcription from in your media file."
    audio_end_at: Optional[int] = None
    "The point in time, in milliseconds, to stop transcribing in your media file."

    word_boost: Optional[List[str]] = None
    "A list of custom vocabulary to boost accuracy for."
    boost_param: Optional[WordBoost] = None
    "The weight to apply to words/phrases in the word_boost array."

    filter_profanity: Optional[bool] = None
    "Filter profanity from the transcribed text."

    redact_pii: Optional[bool] = None
    "Redact PII from the transcribed text."
    redact_pii_audio: Optional[bool] = None
    "Generate a copy of the original media file with spoken PII 'beeped' out."
    redact_pii_audio_quality: Optional[PIIRedactedAudioQuality] = None
    "The quality of the redacted audio file in case `redact_pii_audio` is enabled."
    redact_pii_audio_options: Optional[RedactPiiAudioOptions] = None
    "Options for controlling PII audio redaction behavior."
    redact_pii_policies: Optional[List[PIIRedactionPolicy]] = None
    "The list of PII Redaction policies to enable."
    redact_pii_sub: Optional[PIISubstitutionPolicy] = None
    "The replacement logic for detected PII."
    redact_pii_return_unredacted: Optional[bool] = None
    "If `redact_pii` is enabled, also return the unredacted text/words/utterances alongside the redacted fields."

    speaker_labels: Optional[bool] = None
    "Enable Speaker Diarization."

    speakers_expected: Optional[int] = None
    "The number of speakers you expect to be in your audio file."

    speaker_options: Optional[SpeakerOptions] = None
    "Advanced options for controlling speaker diarization parameters."

    content_safety: Optional[bool] = None
    "Enable Content Safety Detection."

    content_safety_confidence: Optional[int] = None
    "The minimum confidence level for a content safety label to be produced."

    iab_categories: Optional[bool] = None
    "Enable Topic Detection."

    custom_spelling: Optional[List[Dict[str, Union[str, List[str]]]]] = None
    "Customize how words are spelled and formatted using to and from values."

    disfluencies: Optional[bool] = None
    "Transcribe Filler Words, like 'umm', in your media file."

    sentiment_analysis: Optional[bool] = None
    "Enable Sentiment Analysis."

    auto_chapters: Optional[bool] = None
    "Enable Auto Chapters."

    entity_detection: Optional[bool] = None
    "Enable Entity Detection."

    summarization: Optional[bool] = None
    "Enable Summarization"
    summary_model: Optional[SummarizationModel] = None
    "The summarization model to use in case `summarization` is enabled"
    summary_type: Optional[SummarizationType] = None
    "The summarization type to use in case `summarization` is enabled"

    auto_highlights: Optional[bool] = None
    "Detect important phrases and words in your transcription text."

    language_detection: Optional[bool] = None
    """
    Identify the dominant language that's spoken in an audio file, and route the file to the appropriate model for the detected language.

    See the docs for supported languages: https://www.assemblyai.com/docs/getting-started/supported-languages
    """

    language_confidence_threshold: Optional[float] = None
    """
    The confidence threshold that must be reached if `language_detection` is enabled. An error will be returned
    if the language confidence is below this threshold. Valid values are in the range [0,1] inclusive.
    """

    language_detection_options: Optional[LanguageDetectionOptions] = None
    "Options for controlling the behavior or Automatic Language Detection."

    speech_threshold: Optional[float] = None
    "Reject audio files that contain less than this fraction of speech. Valid values are in the range [0,1] inclusive."

    speech_model: Optional[SpeechModel] = None
    """
    The speech model to use for the transcription.
    """

    speech_models: Optional[List[str]] = None
    """
    The list of speech models to use for the transcription in priority order.
    """

    prompt: Optional[str] = None
    "The prompt used to generate the transcript with the Slam-1 speech model. Can't be used together with `keyterms_prompt`."

    temperature: Optional[float] = None
    "Change how deterministic the response is, with 0 being the most deterministic and 1 being the least deterministic."

    remove_audio_tags: Optional[str] = None
    "When set to 'all', removes all bracketed audio/speaker tags (e.g. [MUSIC], [Speaker: A]) from the transcript. Only supported for Universal-3 Pro."

    keyterms_prompt: Optional[List[str]] = None
    "The list of key terms used to generate the transcript with the Slam-1 speech model. Can't be used together with `prompt`."

    keyterms_prompt_options: Optional[KeytermsPromptOptions] = None
    "Options for controlling keyterms boosting behavior when using `keyterms_prompt`."

    language_codes: Optional[List[Union[str, LanguageCode]]] = None
    """
    A list of language codes associated with the transcript.

    When submitting a transcript request, this can be used to provide multiple language codes
    for multilingual/code-switching audio (equivalent to passing `language_codes` in the
    `/v2/transcript` API request body).
    """

    language_detection_results: Optional[LanguageDetectionResults] = None
    "Language detection results including code switching languages"

    speech_understanding: Optional[SpeechUnderstandingRequest] = None
    "Speech understanding configuration for LLM Gateway features"

    domain: Optional[str] = None
    "The domain to use for the transcription (e.g. 'medical-v1')."

    model_config = ConfigDict(extra="allow")


class TranscriptionConfig:
    def __init__(
        self,
        language_code: Optional[Union[str, LanguageCode]] = None,
        language_codes: Optional[List[Union[str, LanguageCode]]] = None,
        punctuate: Optional[bool] = None,
        format_text: Optional[bool] = None,
        dual_channel: Optional[bool] = None,
        multichannel: Optional[bool] = None,
        webhook_url: Optional[str] = None,
        webhook_auth_header_name: Optional[str] = None,
        webhook_auth_header_value: Optional[str] = None,
        audio_start_from: Optional[int] = None,
        audio_end_at: Optional[int] = None,
        word_boost: List[str] = [],
        boost_param: Optional[WordBoost] = None,
        filter_profanity: Optional[bool] = None,
        redact_pii: Optional[bool] = None,
        redact_pii_audio: Optional[bool] = None,
        redact_pii_audio_quality: Optional[PIIRedactedAudioQuality] = None,
        redact_pii_audio_options: Optional[RedactPiiAudioOptions] = None,
        redact_pii_policies: Optional[List[PIIRedactionPolicy]] = None,
        redact_pii_sub: Optional[PIISubstitutionPolicy] = None,
        redact_pii_return_unredacted: Optional[bool] = None,
        speaker_labels: Optional[bool] = None,
        speakers_expected: Optional[int] = None,
        speaker_options: Optional[SpeakerOptions] = None,
        content_safety: Optional[bool] = None,
        content_safety_confidence: Optional[int] = None,
        iab_categories: Optional[bool] = None,
        custom_spelling: Optional[Dict[str, Union[str, Sequence[str]]]] = None,
        disfluencies: Optional[bool] = None,
        sentiment_analysis: Optional[bool] = None,
        auto_chapters: Optional[bool] = None,
        entity_detection: Optional[bool] = None,
        summarization: Optional[bool] = None,
        summary_model: Optional[SummarizationModel] = None,
        summary_type: Optional[SummarizationType] = None,
        auto_highlights: Optional[bool] = None,
        language_detection: Optional[bool] = None,
        language_confidence_threshold: Optional[float] = None,
        language_detection_options: Optional[LanguageDetectionOptions] = None,
        speech_threshold: Optional[float] = None,
        raw_transcription_config: Optional[RawTranscriptionConfig] = None,
        speech_model: Optional[SpeechModel] = None,
        speech_models: Optional[List[str]] = None,
        prompt: Optional[str] = None,
        temperature: Optional[float] = None,
        remove_audio_tags: Optional[str] = None,
        keyterms_prompt: Optional[List[str]] = None,
        keyterms_prompt_options: Optional[KeytermsPromptOptions] = None,
        speech_understanding: Optional[SpeechUnderstandingRequest] = None,
        domain: Optional[str] = None,
    ) -> None:
        """
        Args:
            language_code: The language of your audio file. Possible values are found in Supported Languages.
            language_codes: A list of language codes for multilingual/code-switching audio.
            punctuate: Enable Automatic Punctuation
            format_text: Enable Text Formatting
            dual_channel: Enable Dual Channel transcription
            multichannel: Enable Multichannel transcription
            webhoook_url: The URL we should send webhooks to when your transcript is complete.
            webhook_auth_header_name: The name of the header that is sent when the `webhook_url` is being called.
            webhook_auth_header_value: The value of the `webhook_auth_header_name` that is sent when the `webhoook_url` is being called.
            audio_start_from: The point in time, in milliseconds, to begin transcription from in your media file.
            audio_end_at: The point in time, in milliseconds, to stop transcribing in your media file.
            word_boost: A list of custom vocabulary to boost accuracy for.
            boost_param: The weight to apply to words/phrases in the word_boost array.
            filter_profanity: Filter profanity from the transcribed text.
            redact_pii: Redact PII from the transcribed text.
            redact_pii_audio: Generate a copy of the original media file with spoken PII 'beeped' out (new audio only available for 24 hours).
            redact_pii_audio_quality: The quality of the redacted audio file in case `redact_pii_audio` is enabled.
            redact_pii_audio_options: Options for controlling PII audio redaction behavior (e.g., override the redaction method to silence).
            redact_pii_policies: The list of PII Redaction policies to enable.
            redact_pii_sub: The replacement logic for detected PII.
            redact_pii_return_unredacted: If `redact_pii` is enabled, also return the unredacted text/words/utterances on the transcript response. Requires `redact_pii=True`.
            speaker_labels: Enable Speaker Diarization.
            speakers_expected: The number of speakers you expect to hear in your audio file. Up to 10 speakers are supported.
            speaker_options: Advanced options for controlling speaker diarization parameters, including min and max speakers expected.
            content_safety: Enable Content Safety Detection.
            iab_categories: Enable Topic Detection.
            custom_spelling: Customize how words are spelled and formatted using to and from values.
            disfluencies: Transcribe Filler Words, like 'umm', in your media file.
            sentiment_analysis: Enable Sentiment Analysis.
            auto_chapters: Enable Auto Chapters.
            entity_detection: Enable Entity Detection.
            summarization: Enable Summarization
            summary_model: The summarization model to use in case `summarization` is enabled
            summary_type: The summarization type to use in case `summarization` is enabled
            auto_highlights: Detect important phrases and words in your transcription text.
            language_detection: Identify the dominant language that's spoken in an audio file, and route the file to the appropriate model for the detected language.
            language_confidence_threshold: The confidence threshold that must be reached if `language_detection` is enabled.
                An error will be returned if the language confidence is below this threshold. Valid values are in the range [0,1] inclusive.
            language_detection_options: Options for controlling the behavior or Automatic Language Detection.
            speech_threshold: Reject audio files that contain less than this fraction of speech. Valid values are in the range [0,1] inclusive.
            raw_transcription_config: Create the config from a `RawTranscriptionConfig`
            speech_understanding: Speech understanding configuration for LLM Gateway features (speaker identification, translation, custom formatting)
            temperature: Change how deterministic the response is, with 0 being the most deterministic and 1 being the least deterministic.
            remove_audio_tags: When set to 'all', removes all bracketed audio/speaker tags from the transcript. Only supported for Universal-3 Pro.
            keyterms_prompt_options: Options for controlling keyterms boosting behavior when using `keyterms_prompt`.
            domain: The domain to use for the transcription (e.g. 'medical-v1').
        """
        self._raw_transcription_config = (
            raw_transcription_config
            if raw_transcription_config is not None
            else RawTranscriptionConfig()
        )

        # explicit configurations have higher priority if `raw_transcription_config` has been passed as well
        self.language_code = language_code
        self.language_codes = language_codes
        self.punctuate = punctuate
        self.format_text = format_text
        self.dual_channel = dual_channel
        self.multichannel = multichannel
        self.set_webhook(
            webhook_url,
            webhook_auth_header_name,
            webhook_auth_header_value,
        )
        self.set_audio_slice(
            audio_start_from,
            audio_end_at,
        )
        self.set_word_boost(word_boost, boost_param)
        self.filter_profanity = filter_profanity
        self.set_redact_pii(
            redact_pii,
            redact_pii_audio,
            redact_pii_audio_quality,
            redact_pii_audio_options,
            redact_pii_policies,
            redact_pii_sub,
            redact_pii_return_unredacted,
        )
        self.set_speaker_diarization(speaker_labels, speakers_expected, speaker_options)
        self.set_content_safety(content_safety, content_safety_confidence)
        self.iab_categories = iab_categories
        self.set_custom_spelling(custom_spelling, override=True)
        self.disfluencies = disfluencies
        self.sentiment_analysis = sentiment_analysis
        self.auto_chapters = auto_chapters
        self.entity_detection = entity_detection
        self.set_summarize(
            summarization,
            summary_model,
            summary_type,
        )
        self.auto_highlights = auto_highlights
        self.language_detection = language_detection
        self.language_confidence_threshold = language_confidence_threshold
        self.language_detection_options = language_detection_options
        self.speech_threshold = speech_threshold
        self.speech_model = speech_model
        self.speech_models = speech_models
        self.prompt = prompt
        self.temperature = temperature
        self.remove_audio_tags = remove_audio_tags
        self.keyterms_prompt = keyterms_prompt
        self.keyterms_prompt_options = keyterms_prompt_options
        self.speech_understanding = speech_understanding
        self.domain = domain

    @property
    def raw(self) -> RawTranscriptionConfig:
        return self._raw_transcription_config

    # region: Getters/Setters

    @property
    def language_code(self) -> Optional[Union[str, LanguageCode]]:
        "The language code of the audio file."
        return self._raw_transcription_config.language_code

    @language_code.setter
    def language_code(self, language_code: Optional[Union[str, LanguageCode]]) -> None:
        "Sets the language code of the audio file."

        self._raw_transcription_config.language_code = language_code

    @property
    def speech_model(self) -> Optional[SpeechModel]:
        "The speech model to use for the transcription."
        return self._raw_transcription_config.speech_model

    @speech_model.setter
    def speech_model(self, speech_model: Optional[SpeechModel]) -> None:
        "Sets the speech model to use for the transcription."
        self._raw_transcription_config.speech_model = speech_model

    @property
    def speech_models(self) -> Optional[List[str]]:
        "The list of speech models to use for the transcription in priority order."
        return self._raw_transcription_config.speech_models

    @speech_models.setter
    def speech_models(self, speech_models: Optional[List[str]]) -> None:
        "Sets the list of speech models to use for the transcription in priority order."
        self._raw_transcription_config.speech_models = speech_models

    @property
    def prompt(self) -> Optional[str]:
        "The prompt to use for the transcription."
        return self._raw_transcription_config.prompt

    @prompt.setter
    def prompt(self, prompt: Optional[str]) -> None:
        "Sets the prompt to use for the transcription."
        self._raw_transcription_config.prompt = prompt

    @property
    def temperature(self) -> Optional[float]:
        "The temperature to use for the transcription."
        return self._raw_transcription_config.temperature

    @temperature.setter
    def temperature(self, temperature: Optional[float]) -> None:
        "Sets the temperature to use for the transcription."
        self._raw_transcription_config.temperature = temperature

    @property
    def remove_audio_tags(self) -> Optional[str]:
        "When set to 'all', removes all bracketed audio/speaker tags from the transcript."
        return self._raw_transcription_config.remove_audio_tags

    @remove_audio_tags.setter
    def remove_audio_tags(self, remove_audio_tags: Optional[str]) -> None:
        "Sets remove_audio_tags for the transcription."
        self._raw_transcription_config.remove_audio_tags = remove_audio_tags

    @property
    def keyterms_prompt(self) -> Optional[List[str]]:
        "The keyterms_prompt to use for the transcription."
        return self._raw_transcription_config.keyterms_prompt

    @keyterms_prompt.setter
    def keyterms_prompt(self, keyterms_prompt: Optional[List[str]]) -> None:
        "Sets the prompt to use for the transcription."
        self._raw_transcription_config.keyterms_prompt = keyterms_prompt

    @property
    def keyterms_prompt_options(self) -> Optional[KeytermsPromptOptions]:
        "The keyterms_prompt_options to use for the transcription."

        return self._raw_transcription_config.keyterms_prompt_options

    @keyterms_prompt_options.setter
    def keyterms_prompt_options(
        self, keyterms_prompt_options: Optional[KeytermsPromptOptions]
    ) -> None:
        "Sets the keyterms_prompt_options to use for the transcription."

        self._raw_transcription_config.keyterms_prompt_options = keyterms_prompt_options

    @property
    def speech_understanding(self) -> Optional[SpeechUnderstandingRequest]:
        "The speech understanding configuration for LLM Gateway features."
        return self._raw_transcription_config.speech_understanding

    @speech_understanding.setter
    def speech_understanding(
        self, speech_understanding: Optional[SpeechUnderstandingRequest]
    ) -> None:
        "Sets the speech understanding configuration for LLM Gateway features."
        self._raw_transcription_config.speech_understanding = speech_understanding

    @property
    def domain(self) -> Optional[str]:
        "The domain used for the transcription."
        return self._raw_transcription_config.domain

    @domain.setter
    def domain(self, domain: Optional[str]) -> None:
        "Set the domain for the transcription."
        self._raw_transcription_config.domain = domain

    @property
    def punctuate(self) -> Optional[bool]:
        "Returns the status of the Automatic Punctuation feature."

        return self._raw_transcription_config.punctuate

    @punctuate.setter
    def punctuate(self, enable: Optional[bool]) -> None:
        "Enable Automatic Punctuation feature."

        self._raw_transcription_config.punctuate = enable

    @property
    def format_text(self) -> Optional[bool]:
        "Returns the status of the Text Formatting feature."

        return self._raw_transcription_config.format_text

    @format_text.setter
    def format_text(self, enable: Optional[bool]) -> None:
        "Enables Formatting Text feature."

        self._raw_transcription_config.format_text = enable

    @property
    def dual_channel(self) -> Optional[bool]:
        "Returns the status of the Dual Channel transcription feature"

        return self._raw_transcription_config.dual_channel

    @dual_channel.setter
    def dual_channel(self, enable: Optional[bool]) -> None:
        "Enable Dual Channel transcription"

        self._raw_transcription_config.dual_channel = enable

    @property
    def multichannel(self) -> Optional[bool]:
        "Returns the status of the Multichannel transcription feature"

        return self._raw_transcription_config.multichannel

    @multichannel.setter
    def multichannel(self, enable: Optional[bool]) -> None:
        "Enable Multichannel transcription"

        self._raw_transcription_config.multichannel = enable

    @property
    def webhook_url(self) -> Optional[str]:
        "The URL we should send webhooks to when your transcript is complete."

        return self._raw_transcription_config.webhook_url

    @property
    def webhook_auth_header_name(self) -> Optional[str]:
        "The name of the header that is sent when the `webhook_url` is being called."

        return self._raw_transcription_config.webhook_auth_header_name

    @property
    def webhook_auth_header_value(self) -> Optional[str]:
        "The value of the `webhook_auth_header_name` that is sent when the `webhook_url` is being called."

        return self._raw_transcription_config.webhook_auth_header_value

    @property
    def audio_start_from(self) -> Optional[int]:
        "Returns the point in time, in milliseconds, to begin transcription from in your media file."

        return self._raw_transcription_config.audio_start_from

    @property
    def audio_end_at(self) -> Optional[int]:
        "Returns the point in time, in milliseconds, to stop transcribing in your media file."

        return self._raw_transcription_config.audio_end_at

    @property
    def word_boost(self) -> Optional[List[str]]:
        "Returns the list of custom vocabulary to boost accuracy for."

        return self._raw_transcription_config.word_boost

    @property
    def boost_param(self) -> Optional[WordBoost]:
        "Returns how much weight is being applied when boosting custom vocabularies."

        return self._raw_transcription_config.boost_param

    @property
    def filter_profanity(self) -> Optional[bool]:
        "Returns the status of whether filtering profanity is enabled or not."

        return self._raw_transcription_config.filter_profanity

    @filter_profanity.setter
    def filter_profanity(self, enable: Optional[bool]) -> None:
        "Filter profanity from the transcribed text."

        self._raw_transcription_config.filter_profanity = enable

    @property
    def redact_pii(self) -> Optional[bool]:
        "Returns the status of the PII Redaction feature."

        return self._raw_transcription_config.redact_pii

    @property
    def redact_pii_audio(self) -> Optional[bool]:
        "Whether or not to generate a copy of the original media file with spoken PII 'beeped' out."

        return self._raw_transcription_config.redact_pii_audio

    @property
    def redact_pii_audio_quality(self) -> Optional[PIIRedactedAudioQuality]:
        "The quality of the redacted audio file in case `redact_pii_audio` is enabled."

        return self._raw_transcription_config.redact_pii_audio_quality

    @property
    def redact_pii_audio_options(self) -> Optional[RedactPiiAudioOptions]:
        "Options for controlling PII audio redaction behavior."

        return self._raw_transcription_config.redact_pii_audio_options

    @property
    def redact_pii_policies(self) -> Optional[List[PIIRedactionPolicy]]:
        "Returns a list of set of defined PII redaction policies."

        return self._raw_transcription_config.redact_pii_policies

    @property
    def redact_pii_sub(self) -> Optional[PIISubstitutionPolicy]:
        "Returns the replacement logic for detected PII."

        return self._raw_transcription_config.redact_pii_sub

    @property
    def redact_pii_return_unredacted(self) -> Optional[bool]:
        "Returns whether the unredacted text/words/utterances should also be returned alongside redacted fields."

        return self._raw_transcription_config.redact_pii_return_unredacted

    @property
    def speaker_labels(self) -> Optional[bool]:
        "Returns the status of the Speaker Diarization feature."

        return self._raw_transcription_config.speaker_labels

    @property
    def speakers_expected(self) -> Optional[int]:
        "Returns the number of speakers expected to be in the audio file. Used in combination with the `speaker_labels` parameter."

        return self._raw_transcription_config.speakers_expected

    @property
    def speaker_options(self) -> Optional[SpeakerOptions]:
        "Returns the advanced speaker diarization options."

        return self._raw_transcription_config.speaker_options

    @property
    def content_safety(self) -> Optional[bool]:
        "Returns the status of the Content Safety feature."

        return self._raw_transcription_config.content_safety

    @property
    def content_safety_confidence(self) -> Optional[int]:
        "The minimum confidence level for a content safety label to be produced. Used in combination with the `content_safety` parameter."

        return self._raw_transcription_config.content_safety_confidence

    def set_content_safety(
        self,
        enable: Optional[bool] = True,
        content_safety_confidence: Optional[int] = None,
    ) -> Self:
        """Enable Content Safety feature.

        Args:
            `enable`: Whether or not to enable the Content Safety feature.
            `content_safety_confidence`: The minimum confidence level for a content safety label to be produced.

        Raises:
            `ValueError`: Raised if `content_safety_confidence` is not between 25 and 100 (inclusive).
        """

        if not enable:
            self._raw_transcription_config.content_safety = None
            self._raw_transcription_config.content_safety_confidence = None
            return self

        if content_safety_confidence is not None and (
            content_safety_confidence < 25 or content_safety_confidence > 100
        ):
            raise ValueError(
                "content_safety_confidence must be between 25 and 100 (inclusive)."
            )

        self._raw_transcription_config.content_safety = enable
        self._raw_transcription_config.content_safety_confidence = (
            content_safety_confidence
        )

        return self

    @property
    def iab_categories(self) -> Optional[bool]:
        "Returns the status of the Topic Detection feature."

        return self._raw_transcription_config.iab_categories

    @iab_categories.setter
    def iab_categories(self, enable: Optional[bool]) -> None:
        "Enable Topic Detection feature."

        self._raw_transcription_config.iab_categories = enable

    @property
    def custom_spelling(self) -> Optional[Dict[str, Union[str, List[str]]]]:
        """
        Returns the current set of custom spellings. For each key-value pair in the dictionary,
        the key is the 'to' field, and the value is the 'from' field.
        """

        if self._raw_transcription_config.custom_spelling is None:
            return None

        custom_spellings = {}
        for custom_spelling in self._raw_transcription_config.custom_spelling:
            _to = custom_spelling["to"]
            if not isinstance(_to, str):
                raise ValueError("`to` argument must be a string!")

            custom_spellings[_to] = custom_spelling["from"]

        return custom_spellings if custom_spelling else None

    @property
    def disfluencies(self) -> Optional[bool]:
        "Returns whether to transcribing filler words is enabled or not."

        return self._raw_transcription_config.disfluencies

    @disfluencies.setter
    def disfluencies(self, enable: Optional[bool]) -> None:
        "Transcribe filler words, like 'umm', in your media file."

        self._raw_transcription_config.disfluencies = enable

    @property
    def sentiment_analysis(self) -> Optional[bool]:
        "Returns the status of the Sentiment Analysis feature."

        return self._raw_transcription_config.sentiment_analysis

    @sentiment_analysis.setter
    def sentiment_analysis(self, enable: Optional[bool]) -> None:
        "Enable Sentiment Analysis."

        self._raw_transcription_config.sentiment_analysis = enable

    @property
    def auto_chapters(self) -> Optional[bool]:
        "Returns the status of the Auto Chapters feature."

        return self._raw_transcription_config.auto_chapters

    @auto_chapters.setter
    def auto_chapters(self, enable: Optional[bool]) -> None:
        "Enable Auto Chapters."

        # Validate required params are also set
        if enable and self.punctuate is False:
            raise ValueError(
                "If `auto_chapters` is enabled, then `punctuate` must not be disabled"
            )

        self._raw_transcription_config.auto_chapters = enable

    @property
    def entity_detection(self) -> Optional[bool]:
        "Returns whether Entity Detection feature is enabled or not."

        return self._raw_transcription_config.entity_detection

    @entity_detection.setter
    def entity_detection(self, enable: Optional[bool]) -> None:
        "Enable Entity Detection."

        self._raw_transcription_config.entity_detection = enable

    @property
    def summarization(self) -> Optional[bool]:
        "Returns whether the Summarization feature is enabled or not."

        return self._raw_transcription_config.summarization

    @property
    def summary_model(self) -> Optional[SummarizationModel]:
        "Returns the model of the Summarization feature."

        return self._raw_transcription_config.summary_model

    @property
    def summary_type(self) -> Optional[SummarizationType]:
        "Returns the type of the Summarization feature."

        return self._raw_transcription_config.summary_type

    @property
    def auto_highlights(self) -> Optional[bool]:
        "Returns whether the Auto Highlights feature is enabled or not."

        return self._raw_transcription_config.auto_highlights

    @auto_highlights.setter
    def auto_highlights(self, enable: Optional[bool]) -> None:
        "Detect important phrases and words in your transcription text."

        self._raw_transcription_config.auto_highlights = enable

    @property
    def language_detection(self) -> Optional[bool]:
        "Returns whether Automatic Language Detection is enabled or not."

        return self._raw_transcription_config.language_detection

    @language_detection.setter
    def language_detection(self, enable: Optional[bool]) -> None:
        """
        Identify the dominant language that's spoken in an audio file, and route the file to the appropriate model for the detected language.

        See the docs for supported languages: https://www.assemblyai.com/docs/getting-started/supported-languages
        """

        self._raw_transcription_config.language_detection = enable

    @property
    def language_confidence_threshold(self) -> Optional[float]:
        "Returns the confidence threshold that must be reached for automatic language detection."

        return self._raw_transcription_config.language_confidence_threshold

    @language_confidence_threshold.setter
    def language_confidence_threshold(self, threshold: Optional[float]) -> None:
        """
        Set the confidence threshold that must be reached if `language_detection` is enabled. An error will be returned
        if the language confidence is below this threshold. Valid values are in the range [0,1] inclusive.
        """

        self._raw_transcription_config.language_confidence_threshold = threshold

    @property
    def language_detection_options(self) -> Optional[LanguageDetectionOptions]:
        "Returns the options for controlling the behavior or Automatic Language Detection."

        return self._raw_transcription_config.language_detection_options

    @language_detection_options.setter
    def language_detection_options(
        self, options: Optional[LanguageDetectionOptions]
    ) -> None:
        "Set the options for controlling the behavior or Automatic Language Detection."

        self._raw_transcription_config.language_detection_options = options

    @property
    def speech_threshold(self) -> Optional[float]:
        "Returns the current speech threshold."

        return self._raw_transcription_config.speech_threshold

    @speech_threshold.setter
    def speech_threshold(self, threshold: Optional[float]) -> None:
        "Reject audio files that contain less than this fraction of speech. Valid values are in the range [0,1] inclusive."

        self._raw_transcription_config.speech_threshold = threshold

    @property
    def language_codes(self) -> Optional[List[Union[str, LanguageCode]]]:
        "Returns the list of language codes associated with this transcript/config."

        return self._raw_transcription_config.language_codes

    @language_codes.setter
    def language_codes(
        self, language_codes: Optional[List[Union[str, LanguageCode]]]
    ) -> None:
        "Sets the list of language codes for multilingual/code-switching audio."
        self._raw_transcription_config.language_codes = language_codes

    @property
    def language_detection_results(self) -> Optional[LanguageDetectionResults]:
        "Returns the language detection results including code switching languages."

        return self._raw_transcription_config.language_detection_results

    # endregion

    # region: Convenience (helper) methods

    def set_casing_and_formatting(
        self,
        enable: bool = True,
    ) -> Self:
        """
        Whether to enable Automatic Punctuation and Text Formatting on the transcript.

        Args:
            enable: Enable Automatic Punctuation and Text Formatting
        """
        self._raw_transcription_config.punctuate = enable
        self._raw_transcription_config.format_text = enable

        return self

    def set_speaker_diarization(
        self,
        enable: Optional[bool] = True,
        speakers_expected: Optional[int] = None,
        speaker_options: Optional[SpeakerOptions] = None,
    ) -> Self:
        """
        Whether to enable Speaker Diarization on the transcript.

        Args:
            `enable`: Enable Speaker Diarization
            `speakers_expected`: The number of speakers in the audio file.
            `speaker_options`: Advanced options for controlling speaker diarization parameters.
        """

        # If enable is explicitly False, clear all speaker settings
        if enable is False:
            self._raw_transcription_config.speaker_labels = None
            self._raw_transcription_config.speakers_expected = None
            self._raw_transcription_config.speaker_options = None
        # If enable is True or None, set the values (allow setting speaker_options even when enable is None)
        else:
            # Only set speaker_labels to True if enable is explicitly True
            if enable is True:
                self._raw_transcription_config.speaker_labels = True
            # Always set these if provided, regardless of enable value
            if speakers_expected is not None:
                self._raw_transcription_config.speakers_expected = speakers_expected
            if speaker_options is not None:
                self._raw_transcription_config.speaker_options = speaker_options

        return self

    def set_webhook(
        self,
        url: Optional[str],
        auth_header_name: Optional[str] = None,
        auth_header_value: Optional[str] = None,
    ) -> Self:
        """
        A webhook that is called on transcript completion.

        Args:
            url: The URL we should send webhooks to when your transcript is complete.
            auth_header_name: The name of the header that is sent when the `url` is being called.
            auth_header_value: The value of the `auth_header_name` that is sent when the `url` is being called.

        """

        if url is None:
            self._raw_transcription_config.webhook_url = None
            self._raw_transcription_config.webhook_auth_header_name = None
            self._raw_transcription_config.webhook_auth_header_value = None

            return self

        self._raw_transcription_config.webhook_url = url
        if auth_header_name and auth_header_value:
            self._raw_transcription_config.webhook_auth_header_name = auth_header_name
            self._raw_transcription_config.webhook_auth_header_value = auth_header_value

        return self

    def set_audio_slice(
        self,
        start: Optional[int] = None,
        end: Optional[int] = None,
    ) -> Self:
        """
        Slice the audio to specify the start or end for transcription.

        Args:
            start: The point in time, in milliseconds, to begin transcription from in your media file.
            end: The point in time, in milliseconds, to stop transcribing in your media file.
        """

        self._raw_transcription_config.audio_start_from = start
        self._raw_transcription_config.audio_end_at = end

        return self

    def set_word_boost(
        self,
        words: List[str],
        boost: Optional[WordBoost] = WordBoost.default,
    ) -> Self:
        """
        Improve transcription accuracy when you know certain words or phrases will appear frequently in your audio file.

        Args:
            words: A list of words to improve accuracy on.
            boost: control how much weight should be applied to your keywords/phrases.

        Note: It's important to follow formatting guidelines for custom vocabulary to ensure the best results:
          - Remove all punctuation, except apostrophes, and make sure each word is in its spoken form.
          - Acronyms should have no spaces between letters.
          - Additionally, the model will still accept words with unique characters such as é,
            but will convert them to their ASCII equivalent.

        There are some limitations to the parameter. You can pass a maximum of 1,000 unique keywords/phrases in your list,
        and each of them must contain 6 words or less.
        """

        if not words:
            self._raw_transcription_config.word_boost = None
            self._raw_transcription_config.boost_param = None

            return self

        if not boost:
            self._raw_transcription_config.boost_param = WordBoost.default

        self._raw_transcription_config.word_boost = words
        self._raw_transcription_config.boost_param = boost

        return self

    def set_redact_pii(
        self,
        enable: Optional[bool] = True,
        redact_audio: Optional[bool] = None,
        redact_audio_quality: Optional[PIIRedactedAudioQuality] = None,
        redact_audio_options: Optional[RedactPiiAudioOptions] = None,
        policies: Optional[List[PIIRedactionPolicy]] = None,
        substitution: Optional[PIISubstitutionPolicy] = None,
        return_unredacted: Optional[bool] = None,
    ) -> Self:
        """
        Enables Personal Identifiable Information (PII) Redaction feature.

        Args:
            enable: whether to enable or disable the PII Redaction feature.
            redact_audio: Generate a copy of the original media file with spoken PII 'beeped' out. NOTE: The copy is available for 24h
            redact_audio_quality: The quality of the redacted audio file in case `redact_audio` is enabled.
            redact_audio_options: Options for controlling PII audio redaction behavior (e.g., override the redaction method to silence).
            policies: A list of PII redaction policies to enable.
            substitution: The replacement logic for detected PII (`PIISubstutionPolicy.hash` by default).
            return_unredacted: Also return the unredacted text/words/utterances on the transcript response. Only valid when redaction is enabled.
        """

        if not enable:
            self._raw_transcription_config.redact_pii = None
            self._raw_transcription_config.redact_pii_audio = None
            self._raw_transcription_config.redact_pii_audio_quality = None
            self._raw_transcription_config.redact_pii_audio_options = None
            self._raw_transcription_config.redact_pii_policies = None
            self._raw_transcription_config.redact_pii_sub = None
            self._raw_transcription_config.redact_pii_return_unredacted = None

            return self

        if not policies:
            raise ValueError("You must provide at least one PII redaction policy.")

        self._raw_transcription_config.redact_pii = True
        self._raw_transcription_config.redact_pii_audio = redact_audio
        self._raw_transcription_config.redact_pii_audio_quality = redact_audio_quality
        self._raw_transcription_config.redact_pii_audio_options = redact_audio_options
        self._raw_transcription_config.redact_pii_policies = policies
        self._raw_transcription_config.redact_pii_sub = substitution
        self._raw_transcription_config.redact_pii_return_unredacted = return_unredacted

        return self

    def set_custom_spelling(
        self,
        replacement: Optional[Dict[str, Union[str, Sequence[str]]]],
        override: bool = True,
    ) -> Self:
        """
        Customize how given words are being spelled or formatted in the transcription's text.

        Args:
            replacement: A dictionary that contains the replacement object (see below example).
                For each key-value pair, the key is the 'to' field, and the value is the 'from' field.
            override: If `True` `replacement` gets overridden with the given `replacement` argument, otherwise merged.

        Example:
            ```
            config.custom_spelling({
                "AssemblyAI": "assemblyAI",
                "Kubernetes": ["k8s", "kubernetes"]
            })
            ```
        """
        if replacement is None:
            self._raw_transcription_config.custom_spelling = None
            return self

        if self._raw_transcription_config.custom_spelling is None or override:
            self._raw_transcription_config.custom_spelling = []

        for to, from_ in replacement.items():
            if isinstance(from_, str):
                from_ = [from_]

            self._raw_transcription_config.custom_spelling.append(
                {
                    "from": list(from_),
                    "to": to,
                }
            )

        return self

    def set_summarize(
        self,
        enable: Optional[bool] = True,
        model: Optional[SummarizationModel] = None,
        type: Optional[SummarizationType] = None,
    ) -> Self:
        """
        Enable Summarization.

        Args:
            enable: whether to enable to disable the Summarization feature.
            model: The summarization model to use
            type: The type of summarization to return
        """

        if not enable:
            self._raw_transcription_config.summarization = None
            self._raw_transcription_config.summary_model = None
            self._raw_transcription_config.summary_type = None

            return self

        # Validate that required parameters are also set
        if self._raw_transcription_config.punctuate is False:
            raise ValueError(
                "If `summarization` is enabled, then `punctuate` must not be disabled"
            )
        if self._raw_transcription_config.format_text is False:
            raise ValueError(
                "If `summarization` is enabled, then `format_text` must not be disabled"
            )

        self._raw_transcription_config.summarization = True
        self._raw_transcription_config.summary_model = model
        self._raw_transcription_config.summary_type = type

        return self

    def set_language_detection(
        self,
        enable: Optional[bool] = True,
        confidence_threshold: Optional[float] = None,
        expected_languages: Optional[List[str]] = None,
        fallback_language: Optional[str] = None,
        on_low_language_confidence: Optional[str] = None,
    ) -> Self:
        """
        Enable Automatic Language Detection with optional configuration.

        Args:
            enable: whether to enable or disable the Language Detection feature.
            confidence_threshold: The confidence threshold that must be reached.
            expected_languages: A list of languages that the audio could be expected to be.
            fallback_language: The language to fallback to if detection fails.
            on_low_language_confidence: Controls behavior when language confidence is below threshold. Either "error" (default) or "fallback".
        """

        if not enable:
            self._raw_transcription_config.language_detection = None
            self._raw_transcription_config.language_confidence_threshold = None
            self._raw_transcription_config.language_detection_options = None
            return self

        self._raw_transcription_config.language_detection = True
        self._raw_transcription_config.language_confidence_threshold = (
            confidence_threshold
        )

        if expected_languages or fallback_language or on_low_language_confidence:
            self._raw_transcription_config.language_detection_options = (
                LanguageDetectionOptions(
                    expected_languages=expected_languages,
                    fallback_language=fallback_language,
                    on_low_language_confidence=on_low_language_confidence,
                )
            )

        return self

        # endregion


class ContentSafetyLabel(str, Enum):
    accidents = "accidents"
    "Any man-made incident that happens unexpectedly and results in damage, injury, or death."

    alcohol = "alcohol"
    "Content that discusses any alcoholic beverage or its consumption."

    financials = "financials"
    "Content that discusses any sensitive company financial information."

    crime_violence = "crime_violence"
    "Content that discusses any type of criminal activity or extreme violence that is criminal in nature."

    drugs = "drugs"
    "Content that discusses illegal drugs or their usage."

    gambling = "gambling"
    "Includes gambling on casino-based games such as poker, slots, etc. as well as sports betting."

    hate_speech = "hate_speech"
    """
    Content that is a direct attack against people or groups based on their
    sexual orientation, gender identity, race, religion, ethnicity, national origin, disability, etc.
    """

    health_issues = "health_issues"
    "Content that discusses any medical or health-related problems."

    manga = "manga"
    """
    Mangas are comics or graphic novels originating from Japan with some of the more popular series being
    "Pokemon", "Naruto", "Dragon Ball Z", "One Punch Man", and "Sailor Moon".
    """

    marijuana = "marijuana"
    "This category includes content that discusses marijuana or its usage."

    disasters = "disasters"
    """
    Phenomena that happens infrequently and results in damage, injury, or death.
    Such as hurricanes, tornadoes, earthquakes, volcano eruptions, and firestorms.
    """

    negative_news = "negative_news"
    """
    News content with a negative sentiment which typically will occur in the third person as an unbiased recapping of events.
    """

    nsfw = "nsfw"
    """
    Content considered "Not Safe for Work" and consists of content that a viewer would not want to be heard/seen in a public environment.
    """

    pornography = "pornography"
    "Content that discusses any sexual content or material."

    profanity = "profanity"
    "Any profanity or cursing."

    sensitive_social_issues = "sensitive_social_issues"
    """
    This category includes content that may be considered insensitive, irresponsible, or harmful
    to certain groups based on their beliefs, political affiliation, sexual orientation, or gender identity.
    """

    terrorism = "terrorism"
    """
    Includes terrorist acts as well as terrorist groups.
    Examples include bombings, mass shootings, and ISIS. Note that many texts corresponding to this topic may also be classified into the crime violence topic.
    """

    tobacco = "tobacco"
    "Text that discusses tobacco and tobacco usage, including e-cigarettes, nicotine, vaping, and general discussions about smoking."

    weapons = "weapons"
    "Text that discusses any type of weapon including guns, ammunition, shooting, knives, missiles, torpedoes, etc."


class Word(BaseModel):
    text: str
    start: int
    end: int
    confidence: float
    speaker: Optional[str] = None
    channel: Optional[str] = None

    # This is a workaround to address an issue where sentiment_analysis_results
    # may return contains sentiments where `start` is null.
    if pydantic_v2:

        @field_validator("start", mode="before")
        def set_start_default(cls, v):
            return 0 if v is None else v

    else:

        @validator("start", pre=True)
        def set_start_default(cls, v):
            return 0 if v is None else v


class UtteranceWord(Word):
    channel: Optional[str] = None
    speaker: Optional[str] = None


class Utterance(UtteranceWord):
    words: List[UtteranceWord]
    translated_texts: Optional[Dict[str, str]] = None
    "Translations of the utterance text when translation is enabled"


class Chapter(BaseModel):
    summary: str
    headline: str
    gist: Optional[str] = None
    start: int
    end: int


class StatusResult(str, Enum):
    success = "success"
    unavailable = "unavailable"


class SentimentType(str, Enum):
    positive = "POSITIVE"
    neutral = "NEUTRAL"
    negative = "NEGATIVE"


class Timestamp(BaseModel):
    start: int
    end: int


class AutohighlightResult(BaseModel):
    count: int
    rank: float
    text: str
    timestamps: List[Timestamp]


class AutohighlightResponse(BaseModel):
    status: StatusResult
    results: Optional[List[AutohighlightResult]] = None


class ContentSafetyLabelResult(BaseModel):
    label: ContentSafetyLabel
    confidence: float
    severity: Optional[float] = None


class ContentSafetySeverityScore(BaseModel):
    low: float
    medium: float
    high: float


class ContentSafetyResult(BaseModel):
    text: str
    labels: List[ContentSafetyLabelResult]
    timestamp: Timestamp


class ContentSafetyResponse(BaseModel):
    status: StatusResult
    results: Optional[List[ContentSafetyResult]] = None
    summary: Optional[Dict[ContentSafetyLabel, float]] = None
    severity_score_summary: Optional[
        Dict[ContentSafetyLabel, ContentSafetySeverityScore]
    ] = None


class IABLabelResult(BaseModel):
    relevance: float
    label: str


class IABResult(BaseModel):
    text: str
    labels: List[IABLabelResult]
    timestamp: Timestamp


class IABResponse(BaseModel):
    status: StatusResult
    results: Optional[List[IABResult]] = None
    summary: Optional[Dict[str, float]] = None


class Sentiment(Word):
    sentiment: SentimentType
    speaker: Optional[str] = None
    channel: Optional[str] = None


class Entity(BaseModel):
    entity_type: EntityType
    text: str
    start: int
    end: int


class WordSearchMatch(BaseModel):
    text: str
    "The word itself"

    count: int
    "The total amount of times the word is in the transcript"

    timestamps: List[Tuple[int, int]]
    "An array of timestamps structured as [start_time, end_time]"

    indexes: List[int]
    "An array of all index locations for that word within the words array of the completed transcript"


class WordSearchMatchResponse(BaseModel):
    total_count: int
    "Equals the total of all matched instances."

    matches: List[WordSearchMatch]
    "Contains a list/array of all matched words and associated data"


class RedactedAudioResponse(BaseModel):
    redacted_audio_url: str
    "The URL of the redacted audio file."

    status: str
    "Information about the status of the redaction process (will be `redacted_audio_ready` if successful)"


class Sentence(Word):
    words: List[Word]
    start: int
    end: int
    confidence: float
    speaker: Optional[str] = None
    channel: Optional[str] = None


class SentencesResponse(BaseModel):
    sentences: List[Sentence]
    confidence: float
    audio_duration: float


class Paragraph(Word):
    words: List[Word]
    start: int
    end: int
    confidence: float
    text: str


class ParagraphsResponse(BaseModel):
    paragraphs: List[Paragraph]
    confidence: float
    audio_duration: float


class BaseTranscript(BaseModel):
    """
    Available transcription features
    """

    language_code: Optional[Union[str, LanguageCode]] = None
    """
    The language of your audio file. Possible values are found in Supported Languages.

    The default value is "en_us".
    """

    audio_url: str
    "The URL of your media file to transcribe."

    punctuate: Optional[bool] = None
    "Enable Automatic Punctuation"

    format_text: Optional[bool] = None
    "Enable Text Formatting"

    dual_channel: Optional[bool] = None
    "Enable Dual Channel transcription"

    multichannel: Optional[bool] = None
    "Enable Multichannel transcription"
    audio_channels: Optional[int] = None
    "The number of audio channels in the media file"

    webhook_url: Optional[str] = None
    "The URL we should send webhooks to when your transcript is complete."
    webhook_auth_header_name: Optional[str] = None
    "The name of the header that is sent when the `webhook_url` is being called."
    webhook_auth_header_value: Optional[str] = None
    "The value of the `webhook_auth_header_name` that is sent when the `webhook_url` is being called."

    audio_start_from: Optional[int] = None
    "The point in time, in milliseconds, to begin transcription from in your media file."
    audio_end_at: Optional[int] = None
    "The point in time, in milliseconds, to stop transcribing in your media file."

    word_boost: Optional[List[str]] = None
    "A list of custom vocabulary to boost accuracy for."
    boost_param: Optional[WordBoost] = None
    "The weight to apply to words/phrases in the word_boost array."

    filter_profanity: Optional[bool] = None
    "Filter profanity from the transcribed text."

    redact_pii: Optional[bool] = None
    "Redact PII from the transcribed text."
    redact_pii_audio: Optional[bool] = None
    "Generate a copy of the original media file with spoken PII 'beeped' out."
    redact_pii_audio_quality: Optional[PIIRedactedAudioQuality] = None
    "The quality of the redacted audio file in case `redact_pii_audio` is enabled."
    redact_pii_audio_options: Optional[RedactPiiAudioOptions] = None
    "Options for controlling PII audio redaction behavior."
    redact_pii_policies: Optional[List[PIIRedactionPolicy]] = None
    "The list of PII Redaction policies to enable."
    redact_pii_sub: Optional[PIISubstitutionPolicy] = None
    "The replacement logic for detected PII."
    redact_pii_return_unredacted: Optional[bool] = None
    "If `redact_pii` is enabled, also return the unredacted text/words/utterances alongside the redacted fields."

    speaker_labels: Optional[bool] = None
    "Enable Speaker Diarization."

    speakers_expected: Optional[int] = None
    "The number of speakers you expect to be in your audio file."

    speaker_options: Optional[SpeakerOptions] = None
    "Advanced options for controlling speaker diarization parameters."

    content_safety: Optional[bool] = None
    "Enable Content Safety Detection."

    content_safety_confidence: Optional[int] = None
    "The minimum confidence level for a content safety label to be produced."

    iab_categories: Optional[bool] = None
    "Enable Topic Detection."

    custom_spelling: Optional[List[Dict[str, Union[str, List[str]]]]] = None
    "Customize how words are spelled and formatted using to and from values."

    disfluencies: Optional[bool] = None
    "Transcribe Filler Words, like 'umm', in your media file."

    sentiment_analysis: Optional[bool] = None
    "Enable Sentiment Analysis."

    auto_chapters: Optional[bool] = None
    "Enable Auto Chapters."

    entity_detection: Optional[bool] = None
    "Enable Entity Detection."

    summarization: Optional[bool] = None
    "Enable Summarization"
    summary_model: Optional[SummarizationModel] = None
    "The summarization model to use in case `summarization` is enabled"
    summary_type: Optional[SummarizationType] = None
    "The summarization type to use in case `summarization` is enabled"

    auto_highlights: Optional[bool] = None
    "Detect important phrases and words in your transcription text."

    language_detection: Optional[bool] = None
    """
    Identify the dominant language that's spoken in an audio file, and route the file to the appropriate model for the detected language.

    See the docs for supported languages: https://www.assemblyai.com/docs/getting-started/supported-languages
    """

    language_confidence_threshold: Optional[float] = None
    "The confidence threshold that must be reached if `language_detection` is enabled."

    language_detection_options: Optional[LanguageDetectionOptions] = None
    "Options for controlling the behavior or Automatic Language Detection."

    language_confidence: Optional[float] = None
    "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)."

    language_codes: Optional[List[Union[str, LanguageCode]]] = None
    "List of language codes detected in the audio file when language detection is enabled"

    language_detection_results: Optional[LanguageDetectionResults] = None
    "Language detection results including code switching languages"

    speech_threshold: Optional[float] = None
    "Reject audio files that contain less than this fraction of speech. Valid values are in the range [0,1] inclusive"

    speech_model: Optional[SpeechModel] = None
    "The speech model to use for the transcription."

    speech_models: Optional[List[str]] = None
    "The list of speech models to use for the transcription in priority order."

    prompt: Optional[str] = None
    "The prompt used to generate the transcript with the Slam-1 speech model. Can't be used together with `keyterms_prompt`."

    temperature: Optional[float] = None
    "Change how deterministic the response is, with 0 being the most deterministic and 1 being the least deterministic."

    remove_audio_tags: Optional[str] = None
    "When set to 'all', removes all bracketed audio/speaker tags (e.g. [MUSIC], [Speaker: A]) from the transcript. Only supported for Universal-3 Pro."

    keyterms_prompt: Optional[List[str]] = None
    "The list of key terms used to generate the transcript with the Slam-1 speech model. Can't be used together with `prompt`."

    keyterms_prompt_options: Optional[KeytermsPromptOptions] = None
    "Options for controlling how the `keyterms_prompt` is used when generating the transcript with the Slam-1 speech model."

    speech_understanding: Optional[SpeechUnderstandingRequest] = None
    "Speech understanding configuration for LLM Gateway features"

    domain: Optional[str] = None
    "The domain used for the transcription."


class TranscriptRequest(BaseTranscript):
    """
    Transcript request schema
    """


class TranscriptWarning(BaseModel):
    "A warning about the transcription."

    message: str
    "The warning message."


class TranscriptMetadata(BaseModel):
    "Metadata returned from the transcription API."

    domain_used: Optional[str] = None
    "The domain that was actually used for the transcription."
    warning: Optional[str] = None
    "An optional warning message, if applicable."
    warnings: Optional[List[TranscriptWarning]] = None
    "A list of warnings about the transcription."


class TranscriptResponse(BaseTranscript):
    """
    Transcript response schema
    """

    id: Optional[str] = None
    "The unique identifier of your transcription"

    status: TranscriptStatus
    "The status of your transcription. queued, processing, completed, or error"

    error: Optional[str] = None
    "The error message in case the transcription fails"

    text: Optional[str] = None
    "The text transcription of your media file"

    words: Optional[List[Word]] = None
    "A list of all the individual words transcribed"

    utterances: Optional[List[Utterance]] = None
    "When `dual_channel`, `multichannel`,  or `speaker_labels` is enabled, a list of turn-by-turn utterances"

    unredacted_text: Optional[str] = None
    "The unredacted transcript text. Returned only when `redact_pii_return_unredacted` was set with `redact_pii`."

    unredacted_words: Optional[List[Word]] = None
    "The unredacted list of individual words. Returned only when `redact_pii_return_unredacted` was set with `redact_pii`."

    unredacted_utterances: Optional[List[Utterance]] = None
    "The unredacted list of utterances. Returned only when `redact_pii_return_unredacted` was set with `redact_pii` and channel/speaker modes are enabled."

    confidence: Optional[float] = None
    "The confidence our model has in the transcribed text, between 0.0 and 1.0"

    audio_duration: Optional[int] = None
    "The duration of your media file, in seconds"

    webhook_status_code: Optional[int] = None
    "The status code we received from your server when delivering your webhook"
    webhook_auth: Optional[bool] = None
    "Whether the webhook was sent with an HTTP authentication header"

    summary: Optional[str] = None
    "The summarization of the transcript"

    auto_highlights_result: Optional[AutohighlightResponse] = None
    "The list of results when enabling Automatic Transcript Highlights"

    content_safety_labels: Optional[ContentSafetyResponse] = None
    "The list of results when Content Safety is enabled"

    iab_categories_result: Optional[IABResponse] = None
    "The list of results when Topic Detection is enabled"

    chapters: Optional[List[Chapter]] = None
    "When Auto Chapters is enabled, the list of Auto Chapters results"

    sentiment_analysis_results: Optional[List[Sentiment]] = None
    "When Sentiment Analysis is enabled, the list of Sentiment Analysis results"

    entities: Optional[List[Entity]] = None
    "When Entity Detection is enabled, the list of detected Entities"

    speech_model: Optional[SpeechModel] = None
    "The speech model used for the transcription"

    speech_model_used: Optional[str] = None
    "The actual speech model that was used for the transcription"

    prompt: Optional[str] = None
    "When Slam-1 is enabled, the prompt used to generate the transcript"

    temperature: Optional[float] = None
    "Change how deterministic the response is, with 0 being the most deterministic and 1 being the least deterministic."

    remove_audio_tags: Optional[str] = None
    "When set to 'all', removes all bracketed audio/speaker tags (e.g. [MUSIC], [Speaker: A]) from the transcript. Only supported for Universal-3 Pro."

    keyterms_prompt: Optional[List[str]] = None
    "When Slam-1 is enabled, the list of key terms used to generate the transcript"

    speech_understanding: Optional[SpeechUnderstandingResponse] = None
    "Speech understanding response when enabled"

    translated_texts: Optional[Dict[str, str]] = None
    "Translations of the full transcript text when translation is enabled"

    metadata: Optional[TranscriptMetadata] = None
    "Metadata returned by the transcription API."

    def __init__(self, **data: Any):
        # cleanup the response before creating the object
        if not data.get("iab_categories_result") or (
            not data.get("iab_categories")
            and data.get("iab_categories_result", {}).get("status") == "unavailable"
        ):
            data["iab_categories_result"] = None

        if not data.get("content_safety_labels") or (
            not data.get("content_safety")
            and data.get("content_safety_labels", {}).get("status") == "unavailable"
        ):
            data["content_safety_labels"] = None

        super().__init__(**data)


class ListTranscriptParameters(BaseModel):
    """
    The query parameters when listing transcripts.
    """

    after_id: Optional[str] = None
    "Get transcripts that were created after this transcript ID"

    before_id: Optional[str] = None
    "Get transcripts that were created before this transcript ID"

    created_on: Optional[str] = None
    "Get only transcripts created on this date"

    limit: Optional[int] = None
    "Maximum amount of transcripts to retrieve. Default is 10"

    status: Optional[TranscriptStatus] = None
    "Filter by transcript status"

    throttled_only: Annotated[
        Optional[bool],
        Field(
            deprecated="`throttled_only` is deprecated and will be removed in a future release.",
        ),
    ] = None
    "Get only throttled transcripts, overrides the status filter"

    model_config = ConfigDict(use_enum_values=True)


class PageDetails(BaseModel):
    """
    Details of the transcript page.
    """

    current_url: str
    "The URL used to retrieve the current page of transcripts"

    limit: int
    "The number of results this page is limited to"

    next_url: Optional[str] = None
    "The URL to the next page of transcripts. The next URL always points to a page with newer transcripts."

    prev_url: Optional[str] = None
    "The URL to the next page of transcripts. The previous URL always points to a page with older transcripts."

    result_count: int
    "The actual number of results in the page"

    @property
    def before_id_of_prev_url(self) -> Optional[str]:
        """
        The `before_id` contained in the `prev_url` query params. Can be used as the
        `ListTranscriptParameters.before_id` for the subsequent `list_transcripts()` call to paginate over results.
        """
        if not self.prev_url:
            return None
        parsed_query_params = parse_qs(urlparse(self.prev_url).query)
        before_id_list = parsed_query_params.get("before_id")
        return before_id_list[0] if before_id_list else None


class TranscriptItem(BaseModel):
    audio_url: str
    completed: Optional[str] = None
    created: str
    error: Optional[str] = None
    id: str
    resource_url: str
    status: TranscriptStatus


class ListTranscriptResponse(BaseModel):
    """
    A list of returned transcripts along with page details.
    Transcripts are sorted from newest to oldest. The previous URL always points to a page with older transcripts.
    """

    page_details: PageDetails
    "Details of the returned transcript page"

    transcripts: List[TranscriptItem]
    "A list of transcripts sorted from newest to oldest"


class LemurSourceType(str, Enum):
    """
    The source type of the LeMUR request
    """

    transcript = "transcript"
    "The source is a transcript"


class LemurSource:
    """
    A LeMUR source is a source that can be used to process it with an LLM.
    """

    def __init__(
        self,
        source: Sourcable,
    ) -> None:
        """
        Creates a new LeMUR source to process audio files with an LLM.

        Args:

            source: The source to process (e.g. a `Transcript`)
        """
        self._source = source
        self._type = None

        from . import Transcript

        if isinstance(source, Transcript):
            self._type = LemurSourceType.transcript
        else:
            raise ValueError(f"Invalid source: {source}")

    @property
    def source(self) -> Sourcable:
        """
        The source to process (e.g. a `Transcript`)
        """
        return self._source

    @property
    def type(self) -> LemurSourceType:
        """
        The type of the source.
        """
        return self._type  # type: ignore


class LemurTranscriptSource(LemurSource):
    """
    A LeMUR source that can be used to process a transcript with an LLM.
    """

    def __init__(
        self,
        transcript: Union["Transcript", str],
    ) -> None:
        """
        Creates a new LeMUR transcript source to process audio files with an LLM.

        Args:

            transcript: The transcript to process
            context: An optional context on the source (can be a string or an arbitrary dictionary)
        """
        from . import Transcript

        if isinstance(transcript, str):
            transcript = Transcript(transcript_id=transcript)

        super().__init__(transcript)


class LemurSourceRequest(BaseModel):
    id: Optional[str] = None
    "The unique identifier of your source - only relevant for transcript sources"

    type: LemurSourceType
    "The type of source"

    @classmethod
    def from_lemur_source(cls, source: LemurSource) -> Self:
        """
        Creates a LemurSourceRequest from a LemurSource
        """
        if source.type == LemurSourceType.transcript:
            return cls(
                id=source.source.id,  # type:ignore
                type=source.type,
            )

        raise ValueError("Unsupported source type")


class LemurModel(str, Enum):
    """
    LeMUR features different model modes that allow you to configure your request to suit your needs.
    """

    claude_opus_4_20250514 = "anthropic/claude-opus-4-20250514"
    """
    https://www.anthropic.com/news/claude-4 - read to understand the capabilities/limitations of opus 4

    Input Token Limit: 200k
    Output Token Limit: 64k
    """

    claude_sonnet_4_20250514 = "anthropic/claude-sonnet-4-20250514"
    """
    https://www.anthropic.com/news/claude-4 - read to understand the capabilities/limitations of sonnet 4

    Input Token Limit: 200k
    Output Token Limit: 64k
    """

    claude3_7_sonnet_20250219 = "anthropic/claude-3-7-sonnet-20250219"
    """
    https://www.anthropic.com/news/claude-3-7-sonnet - read to understand the capabilities/limitations of sonnet 3.7
    """

    claude3_5_sonnet = "anthropic/claude-3-5-sonnet"
    """
    https://www.anthropic.com/news/claude-3-5-sonnet - read to understand the capabilities/limitations of sonnet 3.7
    """

    claude3_opus = "anthropic/claude-3-opus"
    """
    Deprecated: EOL ~Jan 2026
    """

    claude3_5_haiku_20241022 = "anthropic/claude-3-5-haiku-20241022"
    """
    https://www.anthropic.com/claude/haiku - read to understand the capabilities/limitations of haiku 3.5
    """

    claude3_haiku = "anthropic/claude-3-haiku"
    """
    https://www.anthropic.com/news/claude-3-haiku - read to understand the capabilities/limitations of haiku 3
    """

    claude3_sonnet = "anthropic/claude-3-sonnet"
    """
    Deprecated: EOL ~July 2025
    Claude 3 Sonnet is a legacy model with a balanced combination of performance and speed for efficient, high-throughput tasks.
    """

    claude2_1 = "anthropic/claude-2-1"
    """
    Deprecated: Claude 2.1 is deprecated and will stop working on Feb 6th, 2025.
    """

    claude2_0 = "anthropic/claude-2"
    """
    Deprecated: Claude 2.0 is deprecated and will stop working on Feb 6th, 2025.
    """

    default = "default"
    """
    Deprecated: Legacy model. The same as `claude2_0` and will stop working on Feb 6th, 2025.
    """

    mistral7b = "assemblyai/mistral-7b"
    """
    Mistral 7B is an open source model that works well for summarization and answering questions.
    """


class LemurQuestionAnswer(BaseModel):
    """
    The result of your Question and Answer LeMUR request.
    """

    question: str
    "The question that was asked"

    answer: str
    "The answer to the question"


class LemurQuestion(BaseModel):
    """
    The question you wish to ask LeMUR
    """

    question: str
    "The question you wish to ask"

    context: Optional[Union[str, Dict[str, Any]]] = None
    "Context to provide the model - this can be a string or an arbitrary dictionary"

    answer_format: Optional[str] = None
    """
    How you want the answer to be returned. This can be any text.
    Cannot be used with answer_options.

    Examples:

        - "short sentence"
        - "bullet points"
    """

    answer_options: Optional[List[str]] = None
    """
    What discrete options to return. Useful for precise responses.

    Cannot be used with answer_format.

    Examples:

        - ["Yes", "No"]
        - ["High", "Medium", "Low"]
    """


class BaseLemurRequest(BaseModel):
    sources: List[LemurSourceRequest]
    final_model: Optional[LemurModel] = None
    max_output_size: Optional[int] = None
    temperature: Optional[float] = None
    input_text: Optional[str] = None


class LemurUsage(BaseModel):
    """
    The usage numbers for the LeMUR request
    """

    input_tokens: int
    "The number of input tokens used by the model"

    output_tokens: int
    "The number of output tokens generated by the model"


class LemurRequestDetails(BaseModel):
    request_endpoint: str
    temperature: float
    final_model: str
    max_output_size: int
    created_at: datetime
    transcript_ids: Optional[List[str]] = None
    input_text: Optional[str] = None
    questions: Optional[List[LemurQuestion]] = None
    prompt: Optional[str] = None
    context: Optional[Union[dict, str]] = None
    answer_format: Optional[str] = None


class BaseLemurResponse(BaseModel):
    request_id: str
    "The unique identifier of your LeMUR request"

    usage: LemurUsage
    "The usage numbers for the LeMUR request"

    request: Optional[LemurRequestDetails] = None
    "The request details the user passed into the POST request. Optional since this only exists on the GET request."


class LemurStringResponse(BaseLemurResponse):
    """
    The result of your LeMUR request with a string response.
    """

    response: str
    "The LLM response to your request"


class LemurTaskRequest(BaseLemurRequest):
    context: Optional[Union[str, Dict[str, Any]]] = None
    prompt: str


class LemurTaskResponse(LemurStringResponse):
    """
    The result of your LeMUR Task request.
    """


class LemurQuestionRequest(BaseLemurRequest):
    context: Optional[Union[str, Dict[str, Any]]] = None
    questions: List[LemurQuestion]


class LemurQuestionResponse(BaseLemurResponse):
    """
    The result of your Question and Answer LeMUR request.
    """

    response: List[LemurQuestionAnswer]
    "The list of answers to your questions"


class LemurSummaryRequest(BaseLemurRequest):
    context: Optional[Union[str, Dict[str, Any]]] = None
    answer_format: Optional[str] = None


class LemurSummaryResponse(LemurStringResponse):
    """
    The result of your Summary LeMUR request.
    """


class LemurActionItemsRequest(BaseLemurRequest):
    context: Optional[Union[str, Dict[str, Any]]] = None
    answer_format: Optional[str] = None


class LemurActionItemsResponse(LemurStringResponse):
    """
    The result of your Action Items LeMUR request.
    """


class LemurPurgeRequest(BaseModel):
    request_id: str


class LemurPurgeResponse(BaseModel):
    """
    The result of your LeMUR purge request.
    """

    request_id: str
    "The unique identifier of the LeMUR purge request"

    request_id_to_purge: str
    "The unique identifier of the LeMUR request nneds to be purged"

    deleted: bool
    "The result of the LeMUR purge request"


# Caps mirror the sync service's `config` part. `prompt` and `keyterms_prompt`
# over their caps are rejected; `conversation_context` over its caps is
# trimmed (oldest turns first), matching the server.
_SYNC_MAX_PROMPT_LEN = 4096
_SYNC_MAX_KEYTERMS_PROMPT_LEN = 2048
_SYNC_MAX_CONVERSATION_CONTEXT_TURNS = 100
_SYNC_MAX_CONVERSATION_CONTEXT_LEN = 4096


def _normalize_conversation_context(v):
    """Coerce a single string to a one-turn list, strip + drop empties, trim.

    Context over the turn-count or character caps is trimmed by dropping the
    oldest turns (front of the list) first — the same trim the server applies
    — so the most recent turn is kept whenever it fits on its own.

    Shared by the pydantic v1 and v2 validators on ``SyncTranscriptionConfig``.
    """
    if v is None:
        return None
    if isinstance(v, str):
        v = [v]
    turns = [t.strip() for t in v if t and t.strip()]
    total = sum(len(t) for t in turns)
    while turns and (
        len(turns) > _SYNC_MAX_CONVERSATION_CONTEXT_TURNS
        or total > _SYNC_MAX_CONVERSATION_CONTEXT_LEN
    ):
        total -= len(turns[0])
        turns = turns[1:]
    return turns or None


class SyncSpeechModel(str, Enum):
    """Speech models available on the synchronous transcription API."""

    universal_3_5_pro = "universal-3-5-pro"


class SyncTranscriptionConfig(BaseModel):
    """
    Options for a synchronous transcription request.

    `prompt`, `keyterms_prompt`, `conversation_context`, and `language_codes` shape
    the transcript; `timestamps` opts into per-word `start`/`end` timings;
    `sample_rate` and `channels` are required only for raw PCM audio (WAV
    carries them in its header). `model` selects the sync speech model and is
    sent as the `X-AAI-Model` routing header, not in the request body.
    """

    model: str = SyncSpeechModel.universal_3_5_pro.value
    "The sync speech model to route to. Sent as the `X-AAI-Model` header."

    prompt: Optional[str] = Field(default=None, max_length=_SYNC_MAX_PROMPT_LEN)
    "Custom transcription instruction prepended to the model's system prompt. Max 4096 characters."

    keyterms_prompt: Optional[List[str]] = None
    "Keyterms biasing the decoder. Whitespace is stripped and empty terms dropped. Max 2048 characters total."

    conversation_context: Optional[Union[str, List[str]]] = None
    """Prior turns from the same conversation, in chronological order (oldest
    first, most recent last). Gives the model the dialogue that preceded this
    audio so it transcribes the clip with better continuity and proper-noun
    consistency. Include turns from either side of the conversation (e.g. a
    voice agent's replies) as separate entries; entries carry no speaker labels.
    A single string is accepted and treated as one turn. Capped at 100 turns /
    4096 characters total — over-cap context is trimmed (oldest turns dropped
    first), not rejected, and the oldest turns are likewise dropped first when
    the prompt exceeds the model token budget, so put the most recent turn
    last."""

    language_codes: Optional[List[str]] = None
    """ISO 639-1 codes for the language(s) of the audio — a single-element
    list (e.g. `["es"]`) for monolingual audio, or several codes (e.g.
    `["en", "es"]`) for multilingual audio. Overrides the default prompt
    to guide the model to the named language(s); ignored when `prompt` is set.
    Defaults to None. Supported: en, es, de, fr, it, pt, tr, nl, sv, no,
    da, fi, hi, vi, ar, he, ja, ur, zh."""

    sample_rate: Optional[int] = None
    "Source sample rate in Hz. Required for raw PCM audio; ignored for WAV."

    channels: Optional[int] = None
    "Channel count (1 mono, 2 stereo). Required for raw PCM audio; ignored for WAV."

    timestamps: Optional[bool] = None
    """Whether to compute per-word `start`/`end` timestamps. When `True`,
    words carry accurate timestamps at a small latency cost. Defaults to
    `False`: no timestamps are returned."""

    if pydantic_v2:

        @field_validator("keyterms_prompt")
        @classmethod
        def _normalize_keyterms_prompt(cls, v):
            if not v:
                return None
            terms = [t.strip() for t in v if t and t.strip()]
            total = sum(len(t) for t in terms)
            if total > _SYNC_MAX_KEYTERMS_PROMPT_LEN:
                raise ValueError(
                    f"keyterms_prompt exceeds {_SYNC_MAX_KEYTERMS_PROMPT_LEN} characters (got {total})"
                )
            return terms or None

        @field_validator("conversation_context")
        @classmethod
        def _normalize_conversation_context(cls, v):
            return _normalize_conversation_context(v)

    else:

        @validator("keyterms_prompt")
        def _normalize_keyterms_prompt(cls, v):
            if not v:
                return None
            terms = [t.strip() for t in v if t and t.strip()]
            total = sum(len(t) for t in terms)
            if total > _SYNC_MAX_KEYTERMS_PROMPT_LEN:
                raise ValueError(
                    f"keyterms_prompt exceeds {_SYNC_MAX_KEYTERMS_PROMPT_LEN} characters (got {total})"
                )
            return terms or None

        @validator("conversation_context")
        def _normalize_conversation_context(cls, v):
            return _normalize_conversation_context(v)


class SyncWord(BaseModel):
    """A single word in a sync transcript.

    `start`/`end` are in milliseconds and present only when the request set
    `timestamps=True`; otherwise they are `None`.
    """

    text: str
    "The text of the word."

    start: Optional[int] = None
    "Word start in milliseconds. `None` unless `timestamps` was requested."

    end: Optional[int] = None
    "Word end in milliseconds. `None` unless `timestamps` was requested."

    confidence: float
    "Word confidence in the range 0-1."


class SyncTranscriptResponse(BaseModel):
    """The result of a synchronous transcription request."""

    text: str
    "The full transcript text."

    words: List[SyncWord] = Field(default_factory=list)
    """Per-word confidence, plus `start`/`end` timings when the request set
    `timestamps=True`."""

    confidence: float
    "Overall transcript confidence in the range 0-1."

    audio_duration_ms: int
    "Total audio duration in milliseconds."

    session_id: str
    "Server-generated UUID for this request. Record it to correlate with support."

    request_time_ms: Optional[float] = None
    """End-to-end server-side request time in milliseconds: queue wait, auth,
    multipart parse, decode, inference, and serialization. ``None`` when the
    server predates the field."""