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
NVIDIA
STT De Conformer-Transducer Large
Model
NVIDIA
NVIDIA
STT De Conformer-Transducer Large

Conformer-Transducer-Large model for German Automatic Speech Recognition, Trained on German NeMo ASRSET

2 Versions
1.6.0
01/20/2022 9:38 PM UTC425.7 MB
1.5.0Selected
12/01/2021 10:35 PM UTC426.89 MBGPU: V100
MODEL
KeyValue
ARCHITECTUREConformer Transducer
INPUTS16000 KHZ MONO-CHANNEL AUDIO (WAV FILES)
OUTPUTSTRANSCRIBED SPEECH
WER greedy
KeyValue
MCV7.0 DEV4.75
MCV7.0 TEST5.36
MLS DEV3.46
MLS TEST4.19
VOX POPULI DEV11.21
VOX POPULI TEST9.14