#7070th of 88 in VoiceFor voice, speech and music
moss-tts-nano
0.1B multilingual text-to-speech model with voice cloning that runs on CPU
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.7k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Language
- Python
1,977 on 22 Apr 20264,668 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
MOSS-TTS-Nano is a 0.1B-parameter autoregressive TTS model (audio tokenizer plus LLM) that outputs 48 kHz stereo audio in 20 languages. It supports voice cloning from a reference clip and streaming generation. It runs through Python scripts, a packaged CLI, a FastAPI web demo, or an ONNX Runtime CPU path without PyTorch. Fine-tuning code and an Android ONNX example are included.
Who it is for: Developers needing small local multilingual TTS with voice cloning on CPU
Strengths
- Streaming generation runs on a 4-core CPU; ONNX path needs no PyTorch
- Voice cloning from a reference clip, with automatic chunking for long text
- Covers 20 languages; ships fine-tuning code and an Android ONNX example
- Optional CUDA through onnxruntime-gpu; weights on Hugging Face and ModelScope
Weaknesses
- Bundled web app is a local demo; OpenAI-compatible endpoint needs vLLM-Omni
- Install can fail on pynini and WeTextProcessing, requiring manual steps
- No Docker or compose files in the repository
- No tagged releases published
What it needs
- GPU optional
- Needs Python 3.12, PyTorch, ONNX Runtime, pynini, WeTextProcessing
- Models: MOSS-TTS-Nano (0.1B, Hugging Face), MOSS-TTS-Nano-100M-ONNX, MOSS-Audio-Tokenizer-Nano
- port 18083
Also in Voice
See all 88| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 98 out of 100 |
| 2 | handyOffline push-to-talk dictation app that pastes transcripts into any text field | 91 out of 100 |
| 3 | whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally | 90 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 84 out of 100 |
| #5 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 83 out of 100 |