Skip to content

Speaker test fix #132

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
AnkushMalaker merged 2 commits into from
Oct 20, 2025
Merged

Speaker test fix #132

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
202acd9
update default .env.template for speaker service to 0.0.0.0 to allow …
AnkushMalaker Oct 20, 2025
d6a6b18
fix for pyannote 4.0
AnkushMalaker Oct 20, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion extras/speaker-recognition/.env.template
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,8 @@ COMPUTE_MODE=cpu
SIMILARITY_THRESHOLD=0.15

# Service Configuration
SPEAKER_SERVICE_HOST=speaker-service
# SPEAKER_SERVICE_HOST: Interface to bind to (0.0.0.0 = all interfaces, allows cross-network access)
SPEAKER_SERVICE_HOST=0.0.0.0
SPEAKER_SERVICE_PORT=8085
SPEAKER_SERVICE_URL=http://speaker-service:8085

Expand Down
18 changes: 14 additions & 4 deletions extras/speaker-recognition/src/simple_speaker_recognition/core/audio_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,10 +77,20 @@ def diarize(self, path: Path, min_speakers: Optional[int] = None, max_speakers:
kwargs['min_speakers'] = min_speakers
if max_speakers is not None:
kwargs['max_speakers'] = max_speakers

diarization = self.diar(str(path), **kwargs)
logger.info(f"Diarization: {diarization}")


output = self.diar(str(path), **kwargs)
logger.info(f"Diarization output: {output}")

# In pyannote.audio 4.0+, the pipeline returns a DiarizeOutput object
# We need to access .speaker_diarization to get the Annotation object
if hasattr(output, 'speaker_diarization'):
diarization = output.speaker_diarization
logger.info(f"Using speaker_diarization from output (pyannote 4.0+)")
else:
# Fallback for older versions (3.x) that return Annotation directly
diarization = output
logger.info(f"Using output directly as Annotation (pyannote 3.x)")

# Apply PyAnnote's built-in gap filling using support() method with configurable collar
# This fills gaps shorter than collar seconds between segments from same speaker
diarization = diarization.support(collar=collar)
Expand Down