- How it works
- Upload client calls, interviews, podcast episodes or voice memos to get editable speech-to-text, with speaker recognition, AI summaries, chapters and action items, and export to formats like SRT, VTT, DOCX or JSON. Files are uploaded to the service rather than processed on-device.
- What's different
- States plainly that accuracy varies with language, accent, background noise and mic quality instead of oversetting expectations, and commits that uploaded data is never used to train AI models.
- Pricing
- Free to start, no credit card required; paid plans from $25/month for creators, teams and API users.
- Best for
- Creators, teams and developers who need transcripts of calls, interviews or recordings with speaker separation and summaries.