#2828th of 38 in VoiceFor voice, speech and music
omnivoice
Multilingual zero-shot text-to-speech model with voice cloning and voice design
Live demo ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 15.3k
- License
- Apache-2.0
- Last commit
- Aug 2026
- Last release
- Jul 2026
- Language
- Python
17 on 31 Mar 202615,349 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OmniVoice is a diffusion language model-style TTS model covering 600+ languages. It clones a voice from a 3–10 second reference clip, or builds one from attributes such as gender, age, pitch and accent. It ships a Python API, three CLIs (Gradio demo, single and multi-GPU batch inference) and training recipes.
Who it is for: Developers needing multilingual TTS or voice cloning they can run themselves
Strengths
- Covers 600+ languages, with a full list in the docs
- Reported RTF of 0.025; FlashInfer path reaches 0.0115 at batch 8 on H100
- Supports CUDA, Apple Silicon (mps) and Intel Arc (xpu) backends
- Inline tags for non-verbal sounds and pinyin or CMU phoneme pronunciation fixes
Weaknesses
- Voice design trained on Chinese and English only; other languages may be unstable
- Cross-lingual cloning carries the reference audio's accent
- No Docker image or compose file; installed via pip or uv
- FlashInfer speedup is NVIDIA-only; flash_attn unavailable on XPU
What it needs
- GPU optional
- Needs PyTorch, torchaudio, Whisper (auto-transcribes reference audio), FlashInfer (optional)
- Models: k2-fsa/OmniVoice (Hugging Face)
- port 8001
Also in Voice
See all 38| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models | 83 out of 100 |
| 2 | whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription | 77 out of 100 |
| 3 | VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech | 71 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 69 out of 100 |
| #5 | meetilyDesktop app that records, transcribes and summarizes meetings locally | 68 out of 100 |