Model
Speech To Text (STT) model based on QuartzNet for recognizing Italian speech.
Use the NGC CLI to download:
Copied!
1 Version
1.0.0rc1Selected03/10/2021 2:32 AM UTC67.96 MB Copied!
1.0.0rc1Selected
03/10/2021 2:32 AM UTC67.96 MB
Copied!
Accuracy
| Key | Value |
|---|---|
| MCV DEV SET WER (%) | 15.22 |
Model
| Key | Value |
|---|---|
| Architecture | quartznet |
| inputs | 16000 KHZ MONO-CHANNEL AUDIO (WAV FILES) |
| outputs | TRANSCRIPTION TEXT |