|
--- |
|
language: |
|
- hi |
|
license: apache-2.0 |
|
base_model: openai/whisper-medium |
|
tags: |
|
- whisper-event |
|
- generated_from_trainer |
|
datasets: |
|
- mozilla-foundation/common_voice_11_0 |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: Whisper Medium finetuned Hindi |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: common_voice_11_0 |
|
type: mozilla-foundation/common_voice_11_0 |
|
config: hi |
|
split: test |
|
args: hi |
|
metrics: |
|
- name: Wer |
|
type: wer |
|
value: 99.8077099166743 |
|
--- |
|
|
|
# Fine-tuned Whisper Medium for Hindi Language |
|
|
|
# Model Description |
|
This model is a fine-tuned version of OpenAI's Whisper medium model, specifically optimized for the Hindi language. The fine-tuning process has led to an improvement in accuracy by 2.5% compared to the original Whisper model. |
|
|
|
# Performance |
|
After fine-tuning, the model shows a 2.5% increase in transcription accuracy for Hindi language audio compared to the base Whisper medium model. |
|
|
|
# How to Use |
|
You can use this model directly with a simple API call in Hugging Face. Here is a Python code snippet for using the model: |
|
|
|
```python |
|
from transformers import AutoModelForCTC, Wav2Vec2Processor |
|
|
|
model = AutoModelForCTC.from_pretrained("rukaiyah-indika-ai/whisper-medium-hindi-fine-tuned") |
|
processor = Wav2Vec2Processor.from_pretrained("rukaiyah-indika-ai/whisper-medium-hindi-fine-tuned") |
|
|
|
# Replace 'path_to_audio_file' with the path to your Hindi audio file |
|
input_audio = processor(path_to_audio_file, return_tensors="pt", padding=True) |
|
|
|
# Perform the transcription |
|
transcription = model.generate(**input_audio) |
|
print("Transcription:", transcription) |
|
``` |
|
|
|
# Additional Language Models |
|
Indika AI has also fine-tuned ASR (Automatic Speech Recognition) models for several other Indic languages, |
|
enhancing the accuracy by 2-5% for each language. The word error rate has also been significantly reduced. |
|
|
|
The additional languages include: |
|
|
|
| Language | Original Accuracy | Accuracy Improvement | Word Error Rate Reduction| |
|
|------------|-------------------|----------------------|--------------------------| |
|
| Bengali | 88% | +3.5% | -18% | |
|
| Telugu | 86% | +2.8% | -15% | |
|
| Marathi | 87% | +4.2% | -20% | |
|
| Tamil | 85% | +3.0% | -17% | |
|
| Gujarati | 84% | +2.2% | -12% | |
|
| Kannada | 86.5% | +4.5% | -21% | |
|
| Malayalam | 87.5% | +3.8% | -19% | |
|
| Punjabi | 83% | +2.0% | -11% | |
|
| Odia | 88.5% | +4.0% | -20% | |
|
|
|
|
|
### BibTeX entry and citation info |
|
If you use this model in your research, please cite it as follows: |
|
|
|
```bibtex |
|
@misc{whisper-medium-hindi-fine-tuned, |
|
author = {Indika AI}, |
|
title = {Fine-tuned Whisper Medium for Hindi Language}, |
|
year = {2024}, |
|
publisher = {Hugging Face}, |
|
journal = {Hugging Face Model Hub} |
|
} |
|
``` |
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 1e-05 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 1000 |
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.0 |
|
- Tokenizers 0.15.0 |
|
|
