Whisper transcription
Build Your Own AI Employee. The complete starter kit for OpenClaw + Hermes Agent. 155 deduplicated skills, drag-and-drop workspace, case studies, install scripts.
npx -y skills add S3YED/appie-kit --skill whisper-transcriptionAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Speech-to-text workflows using Whisper locally or via the OpenAI transcription API. Use when transcribing audio, choosing a Whisper runtime, or producing text, SRT, or JSON transcripts.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.4 KB, 207 tokens by cl100k_base, as published. Nobody here has run it
Whisper Transcription
Overview
Use this umbrella for audio transcription tasks regardless of whether the runtime is local Whisper or the OpenAI API.
Choose a Path
- Local CLI: when you want offline transcription and already have the Whisper binary.
- OpenAI API: when you want hosted transcription, JSON output, or proxy-compatible API access.
Workflow
- Identify the source audio format and desired output (txt, srt, json).
- Pick local or API based on latency, privacy, and environment constraints.
- Set model/language/prompt hints only when they improve the transcript.
- Verify the transcript against the audio length and obvious speaker names.
References
references/path-selection.mdfor runtime choice and output-format notes.
Pitfalls
- Don’t keep separate skills for local and API Whisper when the task class is identical.
- Don’t overfit model choice; prefer the simplest path that meets quality needs.
- Don’t forget to specify output format and destination.