A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.
Python 11.02 k
19 天前
#大语言模型#Open-source industrial-grade ASR models supporting Mandarin, Chinese dialects and English, achieving a new SOTA on public Mandarin ASR benchmarks, while also offering outstanding singing lyrics recogn...
Python 1.05 k
3 个月前
SHALLOW, the first hallucination benchmark for ASR models
Python 0
24 天前