inclusionAI Releases Ming-3.0-tts-0.5B Speech Synthesis Model
Start here
inclusionAI released Ming-3.0-tts-0.5B, a dense 0.5B text-to-speech checkpoint built on a Qwen2.5-0.5B backbone and distributed on Hugging Face under the MIT license. The model supports zero-shot voice cloning, voice design from a natural-language description, 14 inline paralinguistic tags, explicit control of emotion, speed, pitch and volume, and synthesis in Chinese, English, German, Spanish, French, Korean, Japanese, Sichuanese and Cantonese. The Hugging Face repository only distributes the checkpoint and does not provide a standalone inference setup; running the model requires the separate Ming-3.0-tts GitHub repository. The authors state that speaker similarity still trails content accuracy and that stability across multiple paralinguistic events is still being improved.
Generated from reports · updated 9 h ago
Report timeline
Follow the reports to see the event from different sides.
- inclusionAI ModelsinclusionAI releases Ming-3.0-tts-0.5B, a 0.5B multilingual speech synthesis model
inclusionAI released Ming-3.0-tts-0.5B, a dense 0.5B text-to-speech checkpoint built on a Qwen2.5-0.5B backbone and distributed on Hugging Face under the MIT license. It supports zero-shot voice cloning, voice design from a natural-language description, 14 inline paralinguistic tags, explicit emotion, speed, pitch and volume control, and synthesis in Chinese, English, German, Spanish, French, Korean, Japanese plus Sichuanese and Cantonese. Inference requires the separate Ming-3.0-tts GitHub repository; the authors note speaker similarity trails content accuracy and that multi-event paralinguistic stability is still improving.
Heat for this event
- Current, comparable range
- 6
- Peak, comparable range
- 8November 19 18:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.