Open ASR Leaderboard
Compare ASR model WER and speed across languages and datasets
Compare ASR model WER and speed across languages and datasets
Native streaming Chinese/English ASR, 80ms clock
Transcribe audio to text with timestamps and visualization
Transcribe only the target speaker in mixed speech
Shows how to implement streaming multi-speaker ASR
Convert audio files to text
LLM-powered ASR: 31 languages, Chinese dialects, timestamps
Explore Dutch ASR model performance on benchmark datasets
Answer multiple questions in batch locally
Verbatim + intended transcripts with word-level timing
Multilingual speech recognition with Hojo-ASR-Multi-V1
Multilingual speech recognition with frozen Gabor kernels
Southern Kurdish ASR
Transcribe Fon audio or generate Fon speech from text
Private Thai transcription with mixed 4-bit WebGPU inference