A transcription API that goes beyond the Whisper API

If you are looking for a Whisper API, this transcription API returns the same kind of timestamped text and adds speaker labels, summaries and translations on top. Every feature of the app is a JSON endpoint under /api/v1; use the shared TypeScript client or plain HTTP.

• POST /api/v1/transcribe — file (multipart) or URL (JSON)
• GET /api/v1/jobs/:id — poll status and fetch the transcript
• POST /api/v1/insights — summary, chapters, action items
• POST /api/v1/translate — translate segments
• Bearer API key from your account page

curl -X POST https://audiotexta.com/api/v1/transcribe \
  -H "Authorization: Bearer es_YOUR_KEY" \
  -F "file=@meeting.mp3" \
  -F 'options={"language":"en","speakers":true}'

# → { "job": { "id": "tr_…", "status": "queued" } }
curl https://audiotexta.com/api/v1/jobs/tr_… -H "Authorization: Bearer es_YOUR_KEY"

import { AudiotextaClient } from "@echoscribe/shared";
const client = new AudiotextaClient({ baseUrl: "https://audiotexta.com", apiKey: process.env.ECHOSCRIBE_KEY });
const job = await client.transcribeUrl("https://youtu.be/…", { speakers: true });
const done = await client.waitForJob(job.id);
const insights = await client.insights(done, "en");