Create readme
Browse files
README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- Bingsu/zeroth-korean
|
| 4 |
+
language:
|
| 5 |
+
- ko
|
| 6 |
+
metrics:
|
| 7 |
+
- cer
|
| 8 |
+
- wer
|
| 9 |
+
base_model:
|
| 10 |
+
- openai/whisper-large-v3-turbo
|
| 11 |
+
pipeline_tag: automatic-speech-recognition
|
| 12 |
+
Dataset split:
|
| 13 |
+
- test -> 50% validation, 50% test
|
| 14 |
+
- Train set duration: 206 hours 43 minutes
|
| 15 |
+
- Validation set duration: 2 hours 22 minutes
|
| 16 |
+
- Test set duration: 2 hours 22 minutes
|
| 17 |
+
|
| 18 |
+
Results:
|
| 19 |
+
- validation WER: 4.90%
|
| 20 |
+
- validation CER: 1.78%
|
| 21 |
+
- test WER: 4.89
|
| 22 |
+
- test CER: 2.06
|
| 23 |
+
---
|