Getting Started with the Wisprs Transcription API
Wisprs is a voice operations API. Submit a URL — YouTube, TikTok, Instagram, podcast, or a direct audio/video file — and get back a timestamped transcript, SRT subtitles, AI-powered repurposings, and semantic search.
The API is REST-based, uses bearer token authentication, and returns JSON. Jobs are async — you submit a URL, receive a job ID immediately, and either poll for status or configure a webhook to be notified on completion.
What you can do
- Ingest any URL — YouTube, TikTok, Instagram, Spotify, podcast feeds, or any direct audio/video link. No dependency on caption tracks — 100% coverage including Shorts and unlisted content.
- Async transcription — submit a job, get a job ID immediately. A webhook fires when the transcript is ready, or poll
GET /api/v1/jobs/{id}for status. - Export in any format — txt, srt, vtt, json, md, or docx.
- AI repurposing — summaries, chapter markers, quote extraction, Twitter threads, blog posts, show notes.
- Semantic library search — search across all your transcripts by meaning, not just keywords.
- MCP server — plug Wisprs into Claude, Cursor, or any MCP-compatible AI host as a voice operations bridge.
Base URL
https://wisprs.co/api/v1