Generate Subtitles in Another Language

Made from the speech

Translated subtitles here start from what people say. The audio is transcribed, the spoken language is detected automatically, each segment is translated into the language you pick, and the result is written out as timed cues. You don't need an existing subtitle file, and you can't upload one: importing and translating an SRT or VTT you already have isn't supported. If you only have a subtitle file and no recording, a document translator is the right tool for it.

Because the subtitles come from speech, they cover every spoken word and nothing else. Text in the picture, song lyrics you want kept and sound descriptions aren't included.

Choosing the subtitle language

Subtitles can be produced in any of the 21 supported languages. Spanish and Portuguese have variants (Latin America or Spain, Brazil or Portugal) that guide vocabulary and spelling. Czech and Vietnamese are subtitle and transcript languages only, so translated subtitles are the way to reach those viewers. Each run produces one subtitle language; for three languages, run the file three times. Every run also returns transcripts in both the spoken and the subtitle language.

Reading speed in the new language

Viewers read subtitles at a limited speed, and translation changes how much there is to read. German, French and Spanish usually need more words than English for the same idea, so a fast English speaker produces dense German cues. Japanese and Chinese use few characters but each one carries more, so viewers read them more slowly per character. Cue timing follows the original speech segments, so where a cue is too long to read comfortably, split it or shorten it in the SRT. The guides to subtitle readability and subtitle timing give the usual limits per line and per second.

Right-to-left and CJK subtitles

Arabic and Hebrew subtitles are written right to left. The files store them correctly, and YouTube, web players and modern media players display them properly; if you burn them into the picture in an editor, check that the editor supports right-to-left text. Chinese output uses Simplified characters. Japanese has no spaces between words, so line breaks fall between characters; most players handle this, but check long lines on a phone. Right-to-left video localisation covers the rest.

Subtitles or a dub?

Translated subtitles

1 credit per minute (minimum 5 credits per file). Viewers keep the original voices; good for creators, interviews and anything with music.

Dubbed video

50 credits per minute (minimum 2 minutes). Each speaker is voiced in a copy of their own voice in 19 languages; good for training, courses and explainers.

Many publishers do both: subtitles in several languages, and a dub in the one or two languages that bring the most viewers. The subtitles vs dubbing guide walks through the choice.

Several languages from one video

For a YouTube channel, upload each SRT under Subtitles → Add language in YouTube Studio; how to add subtitles to a video has the steps for other platforms. For subtitles in the language that is actually spoken, use the subtitle generator with that language selected.

Using the tool on this page

• The spoken language is detected automatically; you select the output language — the spoken language itself for an untranslated transcript, or another language for a translation. Files can be up to 2 hours long.
• Only the audio is sent for processing — the video stays on your device — and audio and text are deleted within 30 minutes of delivery (how files are protected).
• You can download a timestamped transcript, subtitles (SRT and VTT) and plain text. Every output, step and limit is explained on the subtitle generator page.