feat: add whisper audio transcription service

[ValenColm]

Changes

• Added Whisper-based audio transcription service
• Created app/services/audio_service.py
• Implemented transcription function using openai-whisper
• Added local test script (tests/test_audio.py)
• Updated requirements.txt with whisper dependencies
• Documented setup and execution in README.md
• Added .gitignore for environment and audio files

Scope

This service is responsible ONLY for:

• Transcribing audio into text

Testing

• Tested locally with sample audio file
• Verified transcription output using Whisper model

Requirements

• Python 3.10+
• FFmpeg installed
• Virtual environment recommended

Notes

• Service is isolated and ready to integrate with FastAPI and RabbitMQ
• Designed to plug into AI worker architecture