Transcribe Audio or Video

PythonMedia
POST api.upapi.io/audio-transcribe.post

Transcribe speech from an audio or video URL, with time-coded segments and automatic language detection. Short clips come back in the same call; longer ones return a jobId to poll with audio-transcribe-result. GPU Whisper, up to two hours per file.

audiovideotranscribespeechwhispersubtitles
AuthAPI key
Cost10 units
Timeout25s
Cache43200s

Request schema

FieldTypeDescription
urlreq
stringPublic URL of the audio or video to transcribe (e.g. https://example.com/interview.mp3). Up to 2 hours long.
language
Spoken-language hint as an ISO code (e.g. en). Omit to auto-detect.· default: null

Response schema

FieldTypeDescription
text
Full transcript; null until completed· default: null
jobIdreq
stringPass this to audio-transcribe-result.get to poll
statusreq
stringcompleted once the transcript is ready; processing while it is still running
language
Detected (or forced) language code; null until completed· default: null
progress
integerProgress percentage, 0-100· default: 0
segments
Time-coded segments; null until completed· default: null
durationSec
Audio duration in seconds; null until completed· default: null
pollAfterSec
Seconds to wait before polling again; null once completed· default: null

Try it

Try it

Public URL of the audio or video to transcribe (e.g. https://example.com/interview.mp3). Up to 2 hours long.

Provide a publicly reachable URL — the operation fetches it directly.

Spoken-language hint as an ISO code (e.g. en). Omit to auto-detect.

Run this endpoint live

Jump in as a guest — no signup — and run audio-transcribe.post with your own inputs. Create a free account anytime to get your own API key.

Sign in

Code examples

curl -X POST https://api.upapi.io/audio-transcribe.post \
  -H 'Content-Type: application/json' \
  -H 'x-api-key: YOUR_API_KEY' \
  -d '{ "url": "https://example.com/interview.mp3" }'

Developer tools

OpenAPI specImport the spec or collection into Postman, Insomnia, or any codegen tool.