Commit
•
d67d5c5
1
Parent(s):
b8b7fc1
Update README.md
Browse files
README.md
CHANGED
|
@@ -27,8 +27,18 @@ library_name: transformers
|
|
| 27 |
|
| 28 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) fine-tuned on [IMASc dataset](https://www.kaggle.com/datasets/thennal/imasc).
|
| 29 |
|
|
|
|
|
|
|
| 30 |
IMaSC is a Malayalam text and speech corpus made available by ICFOSS for the purpose of developing speech technology for Malayalam, particularly text-to-speech. The corpus contains 34,473 text-audio pairs of Malayalam sentences spoken by 8 speakers, totalling in approximately 50 hours of audio.
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
The fine-tuned model on evaluating in the following dataset:
|
| 33 |
|
| 34 |
**In Mozilla CommonVoice 11.0 dataset (Malayalam subset):**
|
|
|
|
| 27 |
|
| 28 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) fine-tuned on [IMASc dataset](https://www.kaggle.com/datasets/thennal/imasc).
|
| 29 |
|
| 30 |
+
## About Dataset
|
| 31 |
+
|
| 32 |
IMaSC is a Malayalam text and speech corpus made available by ICFOSS for the purpose of developing speech technology for Malayalam, particularly text-to-speech. The corpus contains 34,473 text-audio pairs of Malayalam sentences spoken by 8 speakers, totalling in approximately 50 hours of audio.
|
| 33 |
|
| 34 |
+
## Training
|
| 35 |
+
|
| 36 |
+
- GPUs used: T4 - 16 GB
|
| 37 |
+
|
| 38 |
+
- Training Time: 14 hours
|
| 39 |
+
|
| 40 |
+
## Evaluation
|
| 41 |
+
|
| 42 |
The fine-tuned model on evaluating in the following dataset:
|
| 43 |
|
| 44 |
**In Mozilla CommonVoice 11.0 dataset (Malayalam subset):**
|