Audio Recorder & Transcriber
When to use: Building recording features: voice memos, interview capture, meeting recording, dictation, and transcription with editing.
npx skills add audio-recorder-transcribergit clone https://github.com/skillhub-community/audio-recorder-transcriber.gitDescription
A full recording → transcription workflow. Recorder: getUserMedia capture with live waveform, pause/resume, elapsed timer, and automatic silence-trimming at the end. PlaybackReview: listen back with the recorded waveform and the ability to trim the start/end before saving. Transcribe: send the blob to a speech-to-text backend (Whisper, assembly, or any adapter) and display the transcript with per-segment timestamps, speaker labels (diarization if available), and inline editing — click a timestamp to jump the audio to that point.
Post-processing: export the transcript as TXT, SRT/VTT subtitles, or structured notes (summary + action items via an LLM pass). Includes a transcript-sync highlighter (the transcript scrolls and highlights in sync with playback), a bilingual view (original + translated side by side, when a translation adapter is configured), and a library of past recordings with full-text search. Handles the mobile gotchas: iOS requires a user gesture to start capture, background-tab recording keeps the timer accurate, and mic permission states are clearly surfaced with retry guidance.
AI Hub
Need a model endpoint?
AI Hub is an OpenAI-compatible endpoint for Cursor, Claude Code, and Agent workflows.
Open AI Hub