speech-to-text
inference-sh6 installs
Turn recordings into timestamped transcripts.
What it does
Converts audio into text transcripts using hosted speech recognition models.
When to use it
Use for transcribing meetings, podcasts, voice notes, and interviews, or creating video captions.
How to use it
Give it an audio or video URL; it returns a JSON transcript with detected language and optional timestamped segments.
What you provide
- an audio or video URL
Uses
Good to know
- edits its own instructions
Latest review
What you need · 3
Requires the inference.sh CLI.
Requires signing in to inference.sh before running transcription.
Access · 2
This skill
inference.sh
Execute
Starts a hosted transcription job
Write
Sends audio URLs to inference.sh
npx skills add belt-sh/cli
Execute
Installs the belt CLI skill