NVIDIA
NVIDIA
STT En ContextNet 256 MLS
Model
NVIDIA
NVIDIA
STT En ContextNet 256 MLS

ContextNet-256 model for English Automatic Speech Recognition, trained on English subset of Multilingual Librispeech Dataset.

  • 1 Version
    1.0.0Selected
    09/16/2021 12:18 AM UTC49.74 MB
    Accuracy
    KeyValue
    Librispeech Dev-Other9.0 %
    Librispeech Test-Other9.2 %
    MLS Dev9.4 %
    MLS Test10.9 %
    Model
    KeyValue
    ARCHITECTUREContextNet
    INPUTS16000 KHZ MONO-CHANNEL AUDIO (WAV FILES)
    OUTPUTSTRANSCRIBED SPEECH