🔊
CosyVoice
Alibaba's multilingual TTS model with voice cloning and instruction-following
Voice AI
CosyVoice is Alibaba's open-source multilingual speech synthesis model that supports voice cloning from a few-second reference audio, cross-lingual voice transfer, and instruction-following for controlling speaking style. It produces high-quality speech in Chinese, English, Japanese, Korean, and other languages with consistent voice characteristics. Developers building multilingual AI applications, localization tools, and voice-enabled products use CosyVoice for its zero-shot cloning capability and instruction control over voice characteristics.
Key Features
- ✓Voice cloning
- ✓Multilingual
- ✓Cross-lingual transfer
- ✓Instruction following
- ✓Zero-shot cloning
#tts#voice-cloning#multilingual#open-source#alibaba
Quick Info
- Category
- Voice AI
- Pricing
- Free