SenseVoice open-source speech understanding model repository: supports ASR, emotion recognition, and audio event detection
QwenAudio/SenseVoice
AI BriefingSenseVoice is a speech-understanding foundation model supporting ASR, language identification, speech emotion recognition, and audio event detection. The released SenseVoiceSmall checkpoint supports Mandarin, Cantonese, English, Japanese, and Korean, and outputs emotion and event labels. It uses a non-autoregressive end-to-end architecture, with inference more than 5 times faster than Whisper-Small and more than 15 times faster than Whisper-Large at comparable parameter sizes.
Source: Tongyi QwenAudio: Original speech project · github.com