AI Audio

Fun-ASR-Realtime Alibaba Cloud

A large streaming real-time speech recognition model launched by Alibaba Qianwen supports real-time recognition of 16 dialects and 30 languages.

Fun-CosyVoice 3.5 FunAudioLLM

The multilingual speech generation model launched by the FunAudioLLM team of Alibaba Tongyi Laboratory supports FreeStyle natural language command control and zero-sample speech cloning.