Qwen Releases a Compact Custom-Voice TTS Model
The new 600-million-parameter model from Alibaba's Qwen team can clone voices from short audio clips for multilingual speech synthesis.
Alibaba's Qwen team has released a new open-source model for text-to-speech, Qwen3-TTS-12Hz-0.6B-CustomVoice. This compact model, with just 600 million parameters, introduces a powerful feature to the open-source audio landscape: custom voice cloning.
With this capability, developers can use a short audio sample to create a digital version of a specific voice. The model can then use this cloned voice to synthesize new speech from text in multiple languages, opening up possibilities for personalized applications, custom voice assistants, and dynamic content creation.
A More Accessible Approach
The model's relatively small size makes it more accessible for researchers and developers to run and fine-tune compared to larger, proprietary systems. The "12Hz" in its name likely points to its internal audio representation, suggesting a design that balances quality with computational efficiency, making it suitable for a wider range of hardware.
This release provides a significant new tool for the open-source AI community. While high-quality TTS models exist, those with built-in, easy-to-use voice cloning are less common. The model is available for download on Hugging Face under the Qwen license, which developers should review for commercial use terms.
Sources
- Visit
Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice
Hugging Face
More from Qwen · Alibaba
All Qwen · Alibaba releases →Qwen-Audio-3.1-ASR Adds Speaker-Aware Transcription
Alibaba's new open audio-language model transcribes and diarizes speech across 30 languages under an Apache 2.0 license.
Qwen-Image-2.1-Turbo targets faster generation
Alibaba's Qwen team ships a distilled, few-step variant of its image model built for speed.
Qwen-Image-2.1 Adds RGBA to Image Generation
Alibaba's Qwen team updates its open image model with transparency support and joint text-to-image and editing capabilities.
More in Text → Speech
All Text → Speech →
Liquid AI's LFM3-Omni-2B brings any-to-any AI to the edge
The 2-billion-parameter model combines vision, audio and speech-to-speech in a package small enough to run on device.
IBM's Granite Switch 4.1 Unifies Chat, Logic, Speech
A single endpoint now routes text conversation, decision-making, and speech synthesis across IBM's Apache-licensed Granite line.
Fish Audio's Drama 3 brings MoE to open TTS
The open-source speech model pairs a 35B mixture-of-experts backbone with voice cloning and directed performance control.
0 comments
No comments yet. Be the first to weigh in.