updated readme
Browse files
README.md
CHANGED
|
@@ -74,7 +74,6 @@ The raw corpora were normalized and cleaned using custom Python scripts. Key ste
|
|
| 74 |
- Removing non-English, unintelligible, or corrupted segments
|
| 75 |
- Normalizing diacritics and fixing broken Unicode characters
|
| 76 |
- Manual filtering of misaligned or low-quality samples
|
| 77 |
-
- Augmenting 50% of training data with offline audio transformations
|
| 78 |
|
| 79 |
## Reproducibility
|
| 80 |
|
|
|
|
| 74 |
- Removing non-English, unintelligible, or corrupted segments
|
| 75 |
- Normalizing diacritics and fixing broken Unicode characters
|
| 76 |
- Manual filtering of misaligned or low-quality samples
|
|
|
|
| 77 |
|
| 78 |
## Reproducibility
|
| 79 |
|