Update README.md
Browse files
README.md
CHANGED
|
@@ -28,14 +28,13 @@ widget:
|
|
| 28 |
|
| 29 |
# Regional bengali text to IPA transcription - byT5-small
|
| 30 |
|
| 31 |
-
<h4 style='background-color: red; color: white; padding: 10px; border-radius: 5px;'><strong>WARNING</strong> The model is being constantly updated. You may see jumps in performance. This will once this warning banner goes away.</h4>
|
| 32 |
|
| 33 |
This is a fine-tuned version of the [google/byt5-small](https://huggingface.co/google/byt5-small) for the task of generating IPA transcriptions from regional bengali text.
|
| 34 |
This was done on the dataset of the competition [“ভাষামূল: মুখের ভাষার খোঁজে“](https://www.kaggle.com/competitions/regipa/overview) by Bengali.AI.
|
| 35 |
|
| 36 |
Model performance:
|
| 37 |
-
- **Word error rate (wer)**: 0.
|
| 38 |
-
- **Char error rate (cer)**: 0.
|
| 39 |
|
| 40 |
|
| 41 |
Supported district tokens:
|
|
|
|
| 28 |
|
| 29 |
# Regional bengali text to IPA transcription - byT5-small
|
| 30 |
|
|
|
|
| 31 |
|
| 32 |
This is a fine-tuned version of the [google/byt5-small](https://huggingface.co/google/byt5-small) for the task of generating IPA transcriptions from regional bengali text.
|
| 33 |
This was done on the dataset of the competition [“ভাষামূল: মুখের ভাষার খোঁজে“](https://www.kaggle.com/competitions/regipa/overview) by Bengali.AI.
|
| 34 |
|
| 35 |
Model performance:
|
| 36 |
+
- **Word error rate (wer)**: 0.0124279344454407
|
| 37 |
+
- **Char error rate (cer)**: 0.00427635805681347
|
| 38 |
|
| 39 |
|
| 40 |
Supported district tokens:
|