Model
NVIDIA
STT En Squeezeformer Large Small LibrispeechSqueezeformer Large model for English Automatic Speech Recognition, Trained with NeMo on LibriSpeech dataset
1.13.0
Copied!
1 Version
1.13.0Selected09/29/2022 9:41 PM UTC902.14 MB Copied!
1.13.0Selected
09/29/2022 9:41 PM UTC902.14 MB
Copied!
Accuracy
| Key | Value |
|---|---|
| LibrispeechTestOther | 5.7 % |
| LibrispeechDevClean | 2.3 % |
| LibrispeechDevOther | 5.7 % |
| LibrispeechTestClean | 2.4 % |
Model
| Key | Value |
|---|---|
| ARCHITECTURE | Squeezeformer |
| INPUTS | 16000 KHZ MONO-CHANNEL AUDIO (WAV FILES) |
| OUTPUTS | TRANSCRIBED SPEECH |