菜单
Transcribe audio files (e.g. Discord .ogg voice messages) using the OpenAI Whisper API. Use when: a voice message or audio attachment needs to be converted to text. NOT for: real-time streaming audio, files >25MB.
Templates, checklists, and automated quality checks for blog posts, email campaigns, landing pages, and brand voice consistency.
Queue health dashboard — scans all role queue directories for a hive and produces a unified JSON health report with per-role metrics, dispatcher heartbeat, failure categorization, and stuck task detection
Mandatory ordered heartbeat gates for hive-supervisor runs
Use this skill to generate well-branded interfaces and assets for HiveWright, either for production or throwaway prototypes/mocks/etc. Contains essential design guidelines, colors, type, fonts, assets, and UI kit components for prototyping.
Quality-doctor diagnosis procedure that gates artifact evidence and enforces the canonical cause-to-action contract before JSON output
On-page SEO analysis — keyword density, meta tags, heading structure, content gaps, competitor comparison
| name | voice-transcription |
| description | Transcribe audio files (e.g. Discord .ogg voice messages) using the OpenAI Whisper API. Use when: a voice message or audio attachment needs to be converted to text. NOT for: real-time streaming audio, files >25MB. |
| homepage | https://platform.openai.com/docs/guides/speech-to-text |
| metadata | {"openclaw":{"emoji":"🎙️","requires":{"bins":["curl","jq"],"env":["OPENAI_API_KEY"]}}} |
Transcribe audio files to text using the OpenAI Whisper API.
USE this skill when:
DON'T use this skill when:
OPENAI_API_KEY is availablebash skills/voice-transcription/transcribe.sh <audio-file-path>
The transcription is printed to stdout. All errors go to stderr.
flac, mp3, mp4, mpeg, mpga, m4a, ogg, opus, wav, webm
Whisper API pricing: $0.006 per minute of audio.
Every call is logged to $HOME/hivewright/logs/voice-transcription-cost.log.
| Exit Code | Cause |
|---|---|
| 1 | No file path argument provided |
| 1 | File not found at given path |
| 1 | Unsupported file format |
| 1 | curl or jq not installed |
| 1 | OPENAI_API_KEY not set |
| 1 | Whisper API returned non-200 response |
| 1 | Empty or null transcription returned |
$HOME/hivewright/logs/voice-transcription-cost.log
Each line: YYYY-MM-DD HH:MM:SS | file=<basename> | size=<bytes>B | model=whisper-1 | cost=~$0.001-0.006