Update README.md
Browse files
README.md
CHANGED
|
@@ -47,17 +47,6 @@ This model is a fine-tuned version of [openai/whisper-small](https://huggingface
|
|
| 47 |
- **Parameters**: ~244M
|
| 48 |
- **License**: Apache 2.0 (following Whisper's license)
|
| 49 |
|
| 50 |
-
## Intended Uses
|
| 51 |
-
|
| 52 |
-
### Direct Use
|
| 53 |
-
|
| 54 |
-
This model is intended for transcribing Azerbaijani speech to text. It can be used for:
|
| 55 |
-
|
| 56 |
-
- Transcription services
|
| 57 |
-
- Subtitling and captioning
|
| 58 |
-
- Voice interfaces
|
| 59 |
-
- Accessibility tools
|
| 60 |
-
- Research applications
|
| 61 |
|
| 62 |
### Out-of-Scope Use
|
| 63 |
|
|
@@ -74,6 +63,7 @@ The model should not be used for:
|
|
| 74 |
The model was fine-tuned on Azerbaijani speech datasets. Training details:
|
| 75 |
|
| 76 |
- **Training checkpoint**: 160
|
|
|
|
| 77 |
- **Audio preprocessing**: 16kHz sampling rate
|
| 78 |
- **Data augmentation**: Applied standard audio augmentation techniques
|
| 79 |
|
|
|
|
| 47 |
- **Parameters**: ~244M
|
| 48 |
- **License**: Apache 2.0 (following Whisper's license)
|
| 49 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
|
| 51 |
### Out-of-Scope Use
|
| 52 |
|
|
|
|
| 63 |
The model was fine-tuned on Azerbaijani speech datasets. Training details:
|
| 64 |
|
| 65 |
- **Training checkpoint**: 160
|
| 66 |
+
- **Traing data**: [common_voice_17_0](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0)
|
| 67 |
- **Audio preprocessing**: 16kHz sampling rate
|
| 68 |
- **Data augmentation**: Applied standard audio augmentation techniques
|
| 69 |
|