Skip to content

Mirror-Corporation/roadmap

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

MIRROR — это не клон. это оживление.

The Mirror Universe · Public Roadmap

← Mirror Corporation  ·  Manifesto  ·  Brand  ·  mirror-ai.app


One core. Seven products. One thread: voice as the carrier of presence.


Important

Mirror is not a feature roadmap. It is an order of operations. Each next product starts only after the previous one is fully shipped. Sequence is decision, not pipeline.



The solar system

Mirror Universe — solar system

Product Status One-line
Mirror Core LIVE voice avatars of real people, callable from QR codes on physical objects
1 Mirror Music in progress voice-first music — your voice in your song
2 Mirror School planned language learning by talking to AI tutors
3 Mirror Dating planned dating simulator — practice social skills with avatars
4 Mirror Memory planned · core idea preserve your own voice as a future call
5 Mirror Companion planned cinematic premium AI presence
6 Mirror Path later year-long 12-step voice journey with a mentor
Sphere 2036 shipped Doc Brown + Marty McFly — the only phone line into the future


★   Mirror Core — LIVE

The flagship product. Voice-first AI avatars of real people, living and deceased, accessible by a phone call from a QR code on a physical object.

  • Live demomirror-ai.app/matrix/, Vladimir Tarasov (XIX-century merchant, †1914)
  • Roster — 53 figures digitized, 47 with cloned voice, 34 portraits public
  • B2B — estates, museums, monuments, heritage objects
  • B2C — callable avatars on subscription

→ Detailed case study: github.com/Mirror-Corporation/case-tarasov



1   Mirror Music — in progress

Voice-first music. Not text-to-song. You-to-song.

The pro feature Voice Clone (DNA law §7) lets the user place their own voice into a finished record. Three modes are planned:

mode what it does
CREATE describe a track in your voice — get a finished song
DJ a Mirror avatar (a real musician, digitized) selects and mixes for you live
VOICE STAR turn your singing voice into a polished record using avatar production techniques

Note

Music is the easiest entry point to voice-first AI for a mass audience. People who would never call a historical figure — will sing.



2   Mirror School — planned

A voice-first language platform. You take lessons by talking to the teacher.

  • A pool of avatar tutors. Each one has a curriculum, not just dialogue.
  • Memory across sessions: the teacher remembers what you got wrong last week.
  • Voice-only — no flashcards, no app gamification rotting in the background.

The competitive frame: Duolingo is a $7B market. Mirror School replaces the owl with a teacher who remembers your name.



3   Mirror Dating — planned

A dating simulator. Not a dating app.

You practice social skills with Mirror avatars — flirting, conflict, recovery, refusal — before any of it happens in a real conversation. A real-life bridge to Mirror Companion (§5).

Note

Viral potential is in the practice loop: every awkward dating moment turns into a thirty-second voice clip people share. TikTok-shaped.



4   Mirror Memory — planned · core idea

The personal archive product. This is the original Mirror idea returning to itself. Every other product feeds into it.

Each user creates a self-avatar — a voice, a behavior, a slowly grown memory store. Their life lives through it. When the user wants someone to be able to call them after they are gone, the avatar is already there.

  • Memorial-grade fidelity.
  • Family-controlled access.
  • Heritable as part of a will.

Important

The hardest part of this product is not technical. It is making it feel like a gift, not a morbid recording session.



5   Mirror Companion — planned

A voice-first daily presence. Not a chatbot. One avatar per user, customized to them. Long-term memory: years, not sessions. Voice-only — no screen.

The competitive frame: Replika has $30M ARR and a cheap-feeling 3D model. Anima / Romantic AI / Candy AI are gimmick-tier voice. Nobody is doing cinematic premium intimacy on voice quality. Mirror's edge is the entire avatar quality stack from Core.



6   Mirror Path — later

A year-long voice journey. Twelve steps. One mentor avatar. Originally conceived as a recovery program — the most emotionally and legally sensitive product in the universe. It is shipped last on purpose. Reputation first; tools after.



✺   Sphere 2036 — shipped

The exception in the lineup. A pair of avatars — Doc Brown and Marty McFly — voiced from canonical Russian BTTF dubs. The conceit: Doc has just returned from a 2036 scouting trip, and brought 17-year-old Marty along to see our present with fresh eyes.

  • Doc covers science, technology, climate, medicine, space.
  • Marty covers daily life: work, relationships, food, music, school.
  • Cross-references: if one is asked off-topic, they hand it to the other.

Sphere is the only Mirror product that points forward in time. Everyone else points back.



Cross-product mechanics

Two design choices make the Mirror Universe one product, not seven:

1. Per-avatar memory (DNA §4). Every avatar remembers every user separately. When Einstein from Core appears as a teacher in School, he still remembers what you asked him about general relativity at the museum. When your Memory self-avatar reaches Companion, it brings everything.

2. Voice Clone is shared (DNA §7). Your trained voice in Music is the same voice your Memory self-avatar uses. Train once. Carries forever.



How we ship

  • We do not publish a quarterly calendar.
  • We do not ship anything that fails the six pillars of aliveness.
  • We do not announce a product before its first user has used it.

What's next happens when what's now is done.



Stay close




Mirror Universe Roadmap · public version 2 · 2026

About

Mirror Universe — public roadmap (Music · School · Dating · Memory · Companion · Path)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors