Fix performance issues: updated requirements with exact versions and added model caching

Update API version to 1.2.0 with ffmpeg support

Add ffmpeg system dependencies to fix audio processing

Fix audio processing libraries for HF Spaces v1.1.0

Update README.md

Fix missing dependencies by simplifying dataset loader for inference

Fix ML training module imports for HF Spaces deployment

Fix PyTorch compatibility issues for HF Spaces deployment

Streamline backend: Remove Whisper/Wav2Vec2 models and dependencies

Initial backend deployment - Flask API with ML models

initial commit