ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
  • NVIDIA
    STT Es Conformer CTC Large
    Model
    NVIDIA
    STT Es Conformer CTC Large

    Conformer-CTC-Large model for Spanish Automatic Speech Recognition, Trained on NeMo ASRSET 2.0

  • 1 Version
    1.8.0Selected
    03/02/2022 2:29 AM UTC1.46 GB
  • Accuracy (beam search + LM, size=128, alpha=0.8, beta=1.5)
    KeyValue
    MCV7.0 dev5.0 %
    MLS test3.6 %
    CALLHOME train20.7 %
    CALLHOME devtest21.6 %
    MLS dev3.6 %
    CALLHOME evltest23.3 %
    Voxpopuli dev5.5 %
    MCV7.0 test5.5 %
    Fisher test17.5 %
    Voxpopuli test6.7 %
    Fisher dev17.5 %
    Accuracy (greedy)
    KeyValue
    MCV7.0 dev6.3 %
    MLS test4.2 %
    CALLHOME train22.4 %
    CALLHOME devtest23.7 %
    MLS dev4.3 %
    CALLHOME evltest25.3 %
    Voxpopuli dev6.1 %
    MCV7.0 test6.9 %
    Fisher test18.5 %
    Voxpopuli test7.5 %
    Fisher dev18.3 %
    Model
    KeyValue
    ArchitectureConformer CTC
    OutputsTranscribed speech
    Inputs16 kHz Mono-channel