Model
—
STT Hr Conformer-CTC LargeConformer-CTC-Large model for Croatian automatic speech recognition, trained on ParlaSpeech-HR v1.0 dataset.
Use the NGC CLI to download:
Copied!
1 Version
1.11.0Selected08/04/2022 10:27 PM UTC464.15 MBCopied!
1.11.0Selected
08/04/2022 10:27 PM UTC464.15 MB
Copied!
Accuracy
| Key | Value |
|---|---|
| ParlaSpeech-HR v1.0 Test WER(%) | 4.70 |
| ParlaSpeech-HR v1.0 Dev WER(%) | 4.43 |
Model
| Key | Value |
|---|---|
| Encoder Dimension | 512 |
| Number of Layers | 18 |
| Architecture | Conformer-CTC-Large |
| Dataset | ParlaSpeech-HR v1.0 |
| Outputs | Text in Croatian |
| Inputs | Speech in Croatian |
| Number of Weights | 121M |