Open-source text-to-speech
Qwen3-TTS
A speech-generation family supporting custom voices, voice design and reference-based cloning with natural-language instructions.
- Voice design
- Voice cloning
- Instruction control
Version & availability
- Project release
- 12Hz family
- Embedded demo
- Qwen3-TTS
- Languages
- 10 languages, including Chinese, English, Japanese, Korean, French and Spanish
The official Space presents the Qwen3-TTS family. Available tasks and model sizes are selected inside the demo.
Last checked:
Try Qwen3-TTS
Demo: Qwen3-TTS · Official Qwen demo
Hosted on Hugging Face. GPU queues, sign-in or usage limits may apply. If the demo does not load here, open it directly using the link above.
License & model terms
The official demo is marked Apache 2.0. Check the selected model's card for its terms.
Read the project documentation ↗