What problem does it solve? Converting recordings, voice memos, podcasts, and meeting captures into text normally requires manual uploads to external services. This Skill transcribes local audio and video files directly through the speech-to-text provider configured in your assistant settings. ## Core Features & Use Cases - Multi-format support: Accepts common audio files (mp3, wav, m4a, flac, ogg) and video files (mp4, mov, mkv, webm), automatically extracting audio from video via ffmpeg. - Provider flexibility: Works with OpenAI Whisper, Deepgram, or Google Gemini, whichever is configured under Speech-to-Text settings. - Large file handling: Automatically splits long recordings into 10-minute chunks and stitches the transcript back together. - Use Case: You have a 45-minute recorded customer interview as an mp4. Point the tool at the file and receive the full transcript without leaving your assistant conversation. ## Quick Start Ask the assistant to transcribe the audio file at an absolute path, for example: transcribe the recording at /tmp/meeting.mp3.
