|
--- |
|
license: mit |
|
datasets: |
|
- thennal/IMaSC |
|
language: |
|
- ml |
|
- en |
|
library_name: transformers |
|
--- |
|
|
|
## Malwhisper-v1-small |
|
|
|
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). |
|
|
|
## About Dataset |
|
|
|
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. |
|
|
|
## Training |
|
|
|
- GPUs used: T4 - 16 GB |
|
|
|
- Training Time: 14 hours |
|
|
|
## Evaluation |
|
|
|
The fine-tuned model on evaluating in the following dataset: |
|
|
|
|
|
**In SMC Malayalam Speech Corpus dataset:** |
|
|
|
WER - 73.56 |
|
|
|
CER - 17.82 |