Video translator for audio and video files

How video translation works

Translation starts with a transcript. Kotosay writes down what is said in the original language, and that text is then translated segment by segment. Each translated line keeps the start and end time of the spoken line, and the translated SRT and VTT come from the same generator as every other subtitle file.

The original voice stays as it is. You get a translated text and subtitles, which suits interviews, lectures, family videos and clips for your own channel.

Can I translate audio to text in English?

Yes. Upload a recording in Spanish, French, Arabic or another of the 90+ languages and translate the transcript into English. The same works the other way round. Kotosay is an audio translator for files, so it suits voice messages, calls and podcasts as much as video. Audio translation follows the same steps as video.

Translate audio to text online, or video to text

To translate audio to text online, upload the recording and pick the target language. To translate video to text, do the same with the video file. Kotosay works as a subtitle translator for your own videos, because each translated line keeps its timing. To translate video subtitles in one step, upload the video itself, and with Single file or Unlimited download the translated SRT or VTT.

What the editor shows after translating

The translated lines take the place of the original lines in the editor, each with its start and end time. Fix a name or a term there and the timing stays as it is.

Kotosay deletes the audio automatically 24 hours after transcribing it. The finished video translation stays in your account until you delete it, so you can come back weeks later and fix a term a colleague pointed out.

Check before you publish

AI translation handles everyday speech well and costs much less than a translator. A few things deserve a second look before you publish.

• Names of people, places and brands, which can come out spelled or translated wrongly.
• Numbers, dates and prices, especially when the speaker says them quickly.
• Idioms and slang, which often have no word-for-word match.
• Formal and informal address, such as tú and usted or tu and vous.

TipRead the lines with names, numbers and idioms first. Their timestamps take you to the right moment in your own copy of the recording, where you can listen once more.

Translated subtitles that are easy to read

Spanish, German and French often need more words than English for the same sentence. A cue that fit well in the original can feel crowded once translated. Play the video with the new SRT in your video player once at normal speed, and where a line flashes past, shorten the wording in the editor. The timing of the cue stays put.

Who translates recordings, and why

Some of the people who upload recordings here for translation:

• Teachers who share a recorded lecture with exchange students.
• Researchers with interviews in a language the rest of the team cannot read.
• Families who want subtitles for a grandparent telling stories in their first language.
• Creators who add a second subtitle language to their own videos.
• Reporters who need to follow a source before a native speaker checks the quotes.

When a human translator fits better

For contracts, medical records and anything with legal weight, work with a certified human translator. You can still use the Kotosay transcript as the source text you hand over.

Common language pairs

Spanish to English, English to Spanish, and French, Portuguese and Arabic to English each have their own page, with notes on accents, regional words and false friends.