media-to-transcript
Convert audio and video into timestamped transcripts with speaker identity resolution on Apple Silicon.
Compatible with
Built for
What it does.
Transcribe podcasts, interviews, meetings, lectures, videos, voice notes, screen recordings, and livestream archives into accurate timestamped transcripts. Runs locally on Apple Silicon using Qwen3-ASR by default, with WhisperKit fallback. Supports any yt-dlp-compatible source and local files.
The skill handles diarization through SpeakerKit, assigns anonymous speaker clusters, then resolves real identities from transcript content, media metadata, and web research. Never treats ASR output as identity proof; verifies every speaker name independently.
Outputs include Markdown, TXT, SRT, JSON, RTTM, and speaker-map files with exact verified names, role labels, or conservative uncertainty markers. Includes run logs and troubleshooting guidance for long media, model downloads, and name corrections.
What’s inside.
Everything you need in the bundle. No tutorials. No assembly.
- 01Qwen3-ASR transcription engine with WhisperKit fallback and optional SpeakerKit diarization
- 02Speaker identity resolution workflow combining transcript analysis, media metadata inspection, and web verification
- 03Multiple output formats: Markdown, plain text, SRT subtitles, JSON cues, RTTM diarization, and speaker-map JSON
- 04Terminology glossary builder to improve domain-specific terms, names, acronyms, and technical vocabulary
- 05Word-level or segment-level timestamps with configurable speaker counts and diarization strategies
- 06Troubleshooting references for long media, model caching, empty transcripts, and boundary handling
- 07Raw engine-native artifacts preserved separately for comparison and audit trails
Drop it in. Press go.
- №01
Download
Buy once, get the file by email and on your dashboard.
- №02
Drop it in
Paste it into your agent. One minute.
- №03
Run it
Your agent picks up the new capability immediately.
- №04
Your agent just got smarter
One more thing off your plate — for every future job.
About the creator.
Things people ask.
$9
media-to-transcript