Skip to content

feat: add PocketTTS engine for CPU-optimized TTS #358

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

Jump to bottom
Open
Robinbinu wants to merge 8 commits into
base: master
Choose a base branch
from
Open

feat: add PocketTTS engine for CPU-optimized TTS #358

Robinbinu wants to merge 8 commits into from

Conversation

@Robinbinu
Copy link
Contributor

@Robinbinu Robinbinu commented Jan 19, 2026

Summary

  • Add PocketTTSEngine - Kyutai Labs' lightweight 100M parameter TTS model
  • CPU-optimized inference (~6x real-time performance)
  • Voice cloning via WAV files
  • ~200ms latency to first audio chunk
  • 8 built-in voices (alba, marius, javert, jean, fantine, cosette, eponine, azelma)

Installation

pip install pocket-tts

Usage

from RealtimeTTS import TextToAudioStream, PocketTTSEngine

engine = PocketTTSEngine(voice="alba")
stream = TextToAudioStream(engine)
stream.feed("Hello world!")
stream.play_async()

Test plan

  • Test with built-in voices
  • Test voice cloning with WAV file
  • Verify CPU performance

@Robinbinu
feat: add WebSocket support for TTS and improve engine initialization…
bd27174 
… logic
@Robinbinu
feat: implement WebSocket client demo for TTS with audio playback and…
c5bb56d 
… file saving
@Robinbinu
feat: enhance TTS UI with mode selector and WebSocket support for aud…
0fa82e5 
…io streaming
@Robinbinu
feat: add WebSocket error handling for unsupported system engine and …
cbf090e 
…improve client feedback
@Robinbinu
chore: refactor engine management to track current engine name for im…
e2c6ce8 
…proved voice retrieval
@Robinbinu
docs: update README to enhance clarity on server components and WebSo…
a32e52f 
…cket support for concurrent users
@Robinbinu
fix: add audio format detection and conversion for non-Kokoro TTS eng…
57bbfcb 
…ines
@Robinbinu
feat: add PocketTTS engine for CPU-optimized TTS
b0b4cb7 
Kyutai Labs' Pocket TTS - lightweight 100M parameter model with:
- CPU-optimized inference (~6x real-time performance)
- Voice cloning via WAV files
- ~200ms latency to first audio chunk
- 8 built-in voices

Install with: pip install pocket-tts
@Robinbinu
Copy link
Contributor Author

Robinbinu commented Jan 19, 2026

Hi @KoljaB,
I have added support for PocketTTS, please review and merge when you are available.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@Robinbinu