nvidia
/

nemotron-speech-streaming-en-0.6b

Automatic Speech Recognition

speech-recognition

cache-aware ASR

Eval Results (legacy)

Model card Files Files and versions

Resources

View closed (8)

Mel-spectrogram boundary artifacts degrade quality in real-time mic streaming — fix inside

#15 opened about 2 months ago by

Are there any more metrics or articles on how this compares to other models for streaming?

#13 opened 2 months ago by

How about a TDT based version instead of RNN-T?

#12 opened 3 months ago by

missing punctuation marks

#11 opened 3 months ago by

Can we expect an ONNX quant?

#6 opened 4 months ago by

MLX version planned?

#3 opened 4 months ago by

Multilingual version planned?

#2 opened 5 months ago by