Silencio-2 开放权重情感语音模型发布
AI 总结
作者发布 Silencio-2,这是一款基于 Kyutai Pocket TTS 微调的 1800 万参数文本转语音模型,支持 24 kHz 流式生成、声音克隆和文本内嵌情绪与声音标签。模型权重和代码已通过 Hugging Face 提供,并配有浏览器 Colab 演示。作者称其可在 Apple M4 处理器上达到约 20 倍实时速度,但说明这只是一次笔记本长文本运行,而非基准测试;发布版本也未进行 WER 测试。
深度见解
Silencio-2 提供了一个体积紧凑、可本地部署且支持情感标签与声音克隆的开放权重 TTS 方案,但其性能和标签稳定性尚无独立验证。
I made this, so treat it as biased. It is free and open, with no paid tier or API behind it.
Silencio-2 is a fine-tune of Kyutai's Pocket TTS 18M that follows tags written inside the text:
What a long day. [sigh] <whisper> Keep your voice down. <whisper> Keep your voice down. <angry> I said stop! Weights and code: https://huggingface.co/BlueRoseLabs/Silencio-2
Try it in your browser, no setup: https://colab.research.google.com/drive/1d6GSiRgXbYyBo91DCMrSr5y2LR_peJKy?usp=sharing
The numbers
#Silencio-2#BlueRose Labs#Kyutai#Pocket TTS#Hugging Face#AudioSet#LibriSpeech