Voice Memo Doc

Transcribe audio files with Whisper on Apple Silicon MLX or NVIDIA CUDA (local or OpenAI-compatible API), then generate markdown documents. Supports high-accuracy 2-pass transcription based on the Courtside Edition paper (arxiv:2602.18966), where the AI agent performs domain and entity analysis. Use for file transcription, STT, speech-to-text, audio-to-document, voice memo documentation, backend speed comparison, or high-accuracy transcription on Mac and CUDA machines.

jkf87 Updated

File contents

jkf87/voice-memo-doc commit df6bb51f3b

Frequently asked questions

npx get-skill add jkf87/voice-memo-doc