Model
Fast Conformer-CTC-Large model for English Automatic Speech Recognition, Trained on NeMo ASRSET
Use the NGC CLI to download:
Copied!
1 Version
1.0.0Selected04/25/2023 8:58 PM UTC441.7 MB Copied!
1.0.0Selected
04/25/2023 8:58 PM UTC441.7 MB
Copied!
Accuracy
| Key | Value |
|---|---|
| LibriSpeech Dev-clean | 1.9 |
| LibriSpeech Test-other | 4.2 |
| LibriSpeech Dev-other | 4.2 |
| LibriSpeech Test-clean | 2.1 |
Model
| Key | Value |
|---|---|
| Encoder Dimension | 512 |
| Number of Layers | 18 |
| Architecture | Fast-Conformer-CTC-Large |
| Dataset | NeMo ASRSET 3.0 |
| Outputs | Text in English |
| Inputs | Speech in English |
| Number of Weights | 115M |