Automatic language detection
The spoken language is detected automatically.
Translate spoken audio into text and bilingual subtitles. Review the original and edit your translation.
How it works
Upload, translate, and review in a dedicated workspace.
01
AAC, FLAC, M4A, MP3, OGG, WAV, WebM · Up to 20 MB · 10 minutes
02
The spoken language is detected automatically.
03
Edit translated segments and export original, translated, or bilingual TXT, SRT, and WebVTT.
Translation features
Model-supported languages, editable captions, and clear retention.
The spoken language is detected automatically.
Review the original, play a segment, and use the pencil to edit its translation.
Audio expires 24 hours after upload. Task data and text results are retained long term. Edited drafts stay in this browser.
Audio translation FAQ
Uploads, languages, output, and retention.
Your audio will be uploaded for cloud transcription and translation.
The selector lists all languages supported by the M2M100 model. Spoken language is detected automatically and must also be supported by the translation model.
The original stays read-only. Edit the translation and export original, translated, or bilingual TXT, SRT, and WebVTT.
Audio expires 24 hours after upload. Task data and text results are retained long term. Edited drafts stay in this browser.
Once upload finishes, processing continues in the cloud. You can refresh or reopen this task URL.
This tool produces translated text and captions. The player uses your original audio for review.