Translate a transcript while you transcribe
Add translate_to, a two-letter ISO 639-1 code, to the request that creates the transcription. wait also waits for the translation. The translation guide covers uploads and MCP.
The transcript is created from the spoken audio or the platform captions as usual. Translation then runs on that text, so the original transcript always exists and stays readable even if the translation is not ready yet.
Only one target language is accepted per request. To get a second language, translate the same transcript again after it completes.
curl -X POST "https://api.transcript1.com/v1/transcriptions?wait=60" \
-H "Authorization: Bearer $TRANSCRIPT1_API_KEY" \
-H "Content-Type: application/json" \
-d '{"url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ", "translate_to": "es"}'Translate an existing transcript
POST the language to the transcription. It answers 202 with the translation status. Poll the transcription, or use a webhook. See Translation for the status codes.
The transcription must be completed first, otherwise the API returns transcript_not_ready. Repeating the request is safe: a translation that is processing or completed is returned as it is, and a failed one is retried.
Webhooks send transcription.translation.completed or transcription.translation.failed, so your application does not have to poll.
curl -X POST "https://api.transcript1.com/v1/transcriptions/tr_8fK2Lm0PqRsTuVwXyZaBcD/translations" \
-H "Authorization: Bearer $TRANSCRIPT1_API_KEY" \
-H "Content-Type: application/json" \
-d '{"language": "en"}'Choose bilingual, translated-only or original subtitles
Add translation=both, only or none to GET /v1/transcriptions/:id or to a text, SRT or VTT download. The default is both.
With both, JSON segments carry a translation field, text puts the translated line under the original, and each SRT or VTT cue has two lines. With only, the translated text replaces the original and the timestamps and speakers stay the same.
The response also has a translation object with the language and a status of processing, completed or failed. Reading another format of a finished transcript does not use more minutes.
| translation | What you get |
|---|---|
| both | Original and translated text together, one pair per segment or cue. |
| only | The translated text in place of the original, with the same timing. |
| none | The original transcript only. |
What translation does and does not do
Translation is machine translation of the transcript text. Review names, technical terms and quotations before you publish them, and keep the original transcript for comparison.
transcript1 does not dub or voice-translate a video. There is no translated audio, and it does not burn subtitles into a video. You receive text and subtitle files that you can use in your own editor or player.
Translation costs no minutes. Transcription minutes apply as usual, and a failed translation neither fails the transcript nor refunds it.