Update README.md
Browse files
README.md
CHANGED
|
@@ -27,19 +27,6 @@ This model is a fine-tuned variant of [openai/whisper-base](https://huggingface.
|
|
| 27 |
|
| 28 |
Explore its capabilities in real-time transcription and multilingual support in the demo space: [Whisper-Base-KhmerV2 Demo](https://huggingface.co/spaces/Vira21/Whisper-Base-KhmerV2).
|
| 29 |
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
## Model Overview
|
| 33 |
-
|
| 34 |
-
- **Base Model**: OpenAI Whisper
|
| 35 |
-
- **Language**: Khmer
|
| 36 |
-
- **Datasets Used**:
|
| 37 |
-
- Google Fleurs
|
| 38 |
-
- OpenSLR
|
| 39 |
-
- Khmer Kheng Info Speech (seanghay/khmer_kheng_info_speech)
|
| 40 |
-
- KM-Speech-Corpus (seanghay/km-speech-corpus)
|
| 41 |
-
- Khmer Grkpp Speech (seanghay/khmer_grkpp_speech)
|
| 42 |
-
- English Dialects (ylacombe/english_dialects)
|
| 43 |
- **Metrics**:
|
| 44 |
- **WER (Word Error Rate)**: 0.4529
|
| 45 |
- **Training Loss**: 0.1012
|
|
|
|
| 27 |
|
| 28 |
Explore its capabilities in real-time transcription and multilingual support in the demo space: [Whisper-Base-KhmerV2 Demo](https://huggingface.co/spaces/Vira21/Whisper-Base-KhmerV2).
|
| 29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
- **Metrics**:
|
| 31 |
- **WER (Word Error Rate)**: 0.4529
|
| 32 |
- **Training Loss**: 0.1012
|