#4343rd of 51 in VoiceFor voice, speech and music
qwen3-tts
Open-weight multilingual text-to-speech models 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
- 13.9k
- License
- Apache-2.0
- Last commit
- Mar 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
Qwen3-TTS is a family of 0.6B and 1.7B text-to-speech models from Qwen, with a Python package (qwen-tts) for custom-voice, voice-design and voice-clone generation. It covers 10 languages, supports streaming with a stated end-to-end latency as low as 97ms, and clones a voice from a 3-second reference clip. The repo also has a local web UI demo, vLLM usage notes and fine-tuning instructions.
Who it is for: Developers adding self-hosted TTS, voice cloning or voice design to apps
Strengths
- Three modes in one family: preset speakers, text-described voices, and 3-second voice cloning
- Covers 10 languages plus dialect voices such as Beijing and Sichuan Chinese
- Streaming and non-streaming from one model; README cites 97ms end-to-end latency
- Apache-2.0 weights on Hugging Face and ModelScope, with fine-tuning and vLLM docs
Weaknesses
- Examples assume a CUDA GPU; CPU-only use is not documented in the README
- Instruction control is only listed for the 1.7B CustomVoice and VoiceDesign models
- Only 9 preset speakers in the CustomVoice models
- No Docker image or release tags; install is via pip in a Python 3.12 environment
What it needs
- GPU required
- Needs PyTorch, FlashAttention 2 (recommended), Hugging Face Transformers, soundfile
- Models: Qwen3-TTS-12Hz-1.7B-VoiceDesign, Qwen3-TTS-12Hz-1.7B-CustomVoice, Qwen3-TTS-12Hz-1.7B-Base, Qwen3-TTS-12Hz-0.6B-CustomVoice, Qwen3-TTS-12Hz-0.6B-Base
- port 8000
Also in Voice
See all 51| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models | 83 out of 100 |
| 2 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 80 out of 100 |
| 3 | whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription | 78 out of 100 |
| #4 | VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech | 71 out of 100 |
| #5 | buzzOffline desktop app for transcribing and translating audio with Whisper | 71 out of 100 |