Indic Whisper Hi Multi Gpu

🚀 IndicWhisper With JAX (more faster)

IndicWhisper is a state - of - the - art speech recognition model fine - tuned on Indian languages, offering pre - trained checkpoints and code for training and evaluation.

🚀 Quick Start

IndicWhisper is a state - of - the - art speech recognition model fine - tuned on Indian languages. This repository contains the code for training and evaluating the model, as well as pre - trained checkpoints for immediate use.

✨ Features

• High Performance: IndicWhisper achieves impressive Word Error Rates (WERs) on various benchmarks for Indian languages, outperforming other publicly available models.
• JAX Mode: Recently added support for JAX mode significantly enhances performance on both TPUs and GPUs, making it the fastest Whisper implementation available.

📚 Documentation

Overview

IndicWhisper achieves impressive Word Error Rates (WERs) on various benchmarks for Indian languages. It outperforms other publicly available models, making it a valuable asset for speech recognition tasks in Indian languages.

Performance on Vistaar Benchmark (Hindi Subset)

💻 Usage Examples

Basic Usage

from whisper_jax import FlaxWhisperForConditionalGeneration, FlaxWhisperPipline
import jax.numpy as jnp

pipeline = FlaxWhisperPipline('parthiv11/indic_whisper_hi_multi_gpu', dtype=jnp.bfloat16)
transcript = pipeline('sample.mp3')

Acknowledgements

We would like to express our gratitude to the following organizations for their support:

• EkStep Foundation for their generous grant, which facilitated the establishment of the Centre for AI4Bharat at IIT Madras.
• The Ministry of Electronics and Information Technology (NLTM) for its grant to support the creation of datasets and models for Indian languages under the Bhashini project.
• The Centre for Development of Advanced Computing, India (C - DAC), for providing access to the Param Siddhi supercomputer for training our models.
• Microsoft for its grant to create datasets, tools, and resources for Indian languages.
• For JAX guide on github

📄 License

IndicWhisper and the associated Vistaar benchmark are MIT - licensed. This license applies to all the fine - tuned language models included in this repository.

Contributors

• Kaushal Bhogale (AI4Bharat)
• Sai Narayan Sundaresan (IITKGP, AI4Bharat)
• Abhigyan Raman (AI4Bharat)
• Tahir Javed (IITM, AI4Bharat)
• Mitesh Khapra (IITM, AI4Bharat, RBCDSAI)
• Pratyush Kumar (Microsoft, AI4Bharat)

🤝 Contributing

We welcome contributions from the community to further improve IndicWhisper. If you have any ideas, bug fixes, or enhancements, please feel free to submit a pull request.

Thank you for your interest in IndicWhisper! We hope it proves to be a valuable tool for your speech recognition needs in Indian languages.