AlexisBalayre/AI-Powered-Meeting-Summarizer

Leverages `whisper.cpp` for efficient local audio transcription and integrates with an `Ollama` server backend for LLM-powered summarization, supporting multiple model variants for both tasks. The Gradio interface enables interactive model selection, optional context injection, and transcript export, while supporting multi-language audio with automatic translation to English. Built on a local-first architecture with FFmpeg for audio preprocessing and downloadable Whisper models (base through large-V3) for flexible performance/accuracy tradeoffs.

144 stars. No commits in the last 6 months.