Gyimah3/Africa-whisper-Finetuned

Framework for seamless fine-tuning and deploying Whisper Model developed to advance Automatic Speech Recognition (ASR): translation and transcription capabilities for African languages African Whisper: Enhanced ASR for African Languages ## 🚀 Enhanced Version This is an **improved and modernized** version of the original African Whisper repository by Kevin Kibe. This enhanced version addresses dependency issues and uses more recent package versions for better stability and performance. ### ✨ Key Improvements - 🔧 **Updated Dependencies**: Resolved compatibility issues with modern package versions - 🚀 **Enhanced Performance**: Optimized for better inference speed and accuracy - 🛡️ **Security Improvements**: Removed hardcoded secrets and enhanced security practices - 📦 **Pre-trained Model**: Includes a fine-tuned model available on Hugging Face - 🐳 **Better Docker Support**: Improved containerization and deployment - 📊 **Enhanced Monitoring**: Better integration with Weights & Biases *Framework for seamless fine-tuning and deploying Whisper Model developed to advance Automatic Speech Recognition (ASR): translation and transcription capabilities for African languages*. ## 🚀 Quick Start ### Using the Pre-trained Model ```bash # Install dependencies pip install -r requirements.txt # Use the fine-tuned model directly python -c " from transformers import WhisperProcessor, WhisperForConditionalGeneration import torch model_name = 'Gyimah3/whisper-small-finetuned' processor = WhisperProcessor.from_pretrained(model_name) model = WhisperForConditionalGeneration.from_pretrained(model_name) print('Model loaded successfully!') " ``` ### Docker Deployment ```bash # Build and run with Docker docker build -t african-whisper . docker run -p 8000:8000 african-whisper ``` ## Features - 🔧 **Fine-Tuning**: Fine-tune the Whisper model on any audio dataset from Huggingface, e.g., Mozilla's Common Voice, Fleurs, LibriSpeech, or your own custom private/public dataset etc - 📊 **Metrics Monitoring**: View training run metrics on Wandb. - 🐳 **Production Deployment**: Seamlessly containerize and deploy the model inference endpoint for real-wor …