Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.08823 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910049045053440 |
|---|---|
| author | Liao, Shijia Wang, Yuxuan Liu, Songting Cheng, Yifan Zhang, Ruoyi Li, Tianyu Li, Shidong Zheng, Yisheng Liu, Xingwei Wang, Qingzheng Zhou, Zhizhuo Liu, Jiahua Chen, Xin Han, Dawei |
| author_facet | Liao, Shijia Wang, Yuxuan Liu, Songting Cheng, Yifan Zhang, Ruoyi Li, Tianyu Li, Shidong Zheng, Yisheng Liu, Xingwei Wang, Qingzheng Zhou, Zhizhuo Liu, Jiahua Chen, Xin Han, Dawei |
| contents | We introduce Fish Audio S2, an open-sourced text-to-speech system featuring multi-speaker, multi-turn generation, and, most importantly, instruction-following control via natural-language descriptions. To scale training, we develop a multi-stage training recipe together with a staged data pipeline covering video captioning and speech captioning, voice-quality assessment, and reward modeling. To push the frontier of open-source TTS, we release our model weights, fine-tuning code, and an SGLang-based inference engine. The inference engine is production-ready for streaming, achieving an RTF of 0.195 and a time-to-first-audio below 100 ms.Our code and weights are available on GitHub (https://github.com/fishaudio/fish-speech) and Hugging Face (https://huggingface.co/fishaudio/s2-pro). We highly encourage readers to visit https://fish.audio to try custom voices. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08823 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Fish Audio S2 Technical Report Liao, Shijia Wang, Yuxuan Liu, Songting Cheng, Yifan Zhang, Ruoyi Li, Tianyu Li, Shidong Zheng, Yisheng Liu, Xingwei Wang, Qingzheng Zhou, Zhizhuo Liu, Jiahua Chen, Xin Han, Dawei Sound Artificial Intelligence Computation and Language We introduce Fish Audio S2, an open-sourced text-to-speech system featuring multi-speaker, multi-turn generation, and, most importantly, instruction-following control via natural-language descriptions. To scale training, we develop a multi-stage training recipe together with a staged data pipeline covering video captioning and speech captioning, voice-quality assessment, and reward modeling. To push the frontier of open-source TTS, we release our model weights, fine-tuning code, and an SGLang-based inference engine. The inference engine is production-ready for streaming, achieving an RTF of 0.195 and a time-to-first-audio below 100 ms.Our code and weights are available on GitHub (https://github.com/fishaudio/fish-speech) and Hugging Face (https://huggingface.co/fishaudio/s2-pro). We highly encourage readers to visit https://fish.audio to try custom voices. |
| title | Fish Audio S2 Technical Report |
| topic | Sound Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2603.08823 |