Skip to main content
NVIDIA
STT De Quartznet15x5
Model
NVIDIA
STT De Quartznet15x5

Speech To Text (STT) model based on QuartzNet for recognizing German speech.

  • 1 Version
    1.0.0rc1Selected
    03/10/2021 2:34 AM UTC67.91 MB
    Accuracy
    KeyValue
    MCV DEV SET WER (%)11.78
    Model
    KeyValue
    ARCHITECTUREQUARTZNET
    INPUTS16000 KHZ MONO-CHANNEL AUDIO (WAV FILES)
    OUTPUTSTRANSCRIPTION TEXT