AI Subtitle Generator

Subtitle files you can use anywhere

SRT

For YouTube, Vimeo, LinkedIn, editors and desktop players — SRT generator

VTT

For HTML5 video and web players — VTT generator

Timestamped TXT

For reading and quoting — transcript with timestamps

Cues follow natural speech segments. Subtitles are delivered as files, not burned into the picture; editors like Premiere Pro or DaVinci Resolve can burn them in if a platform requires it.

Translated subtitles

Choose a target language to get subtitles in that language instead of the spoken one. Each run produces one subtitle track, so run the video again for each extra language.

Adding subtitles to your video

1. YouTube: Studio → Subtitles → Add language → Upload file → With timing.
2. Website: add a <track> element pointing at the VTT file.
3. Editors: import the SRT onto your timeline.

Step-by-step for each platform in how to add subtitles to a video.

How it works

1. Drop in a video. Your browser reads it and pulls out only the audio — the video file never leaves your device.
2. The audio is processed in 30-second chunks: speech is transcribed and the spoken language is detected automatically.
3. Download the results. Audio and text on our servers are deleted after delivery, within 30 minutes at most.

Supported files

• Video: MP4, WebM, MOV, MKV, M4V — whatever your browser can read.
• Length: up to 2 hours per file.
• Languages: speech in 21 languages is recognised automatically; translation and voices cover the same 21.

Privacy

Your video is never uploaded — only its audio is, over HTTPS in small chunks. Audio, transcripts and voice files are deleted after delivery, within 30 minutes at most, and are never used to train models. Your account keeps only the file name, length, languages and credits used. Details in the privacy policy.

Good to know

• No speaker labels — transcripts don't say who is speaking.
• Files only: no links from YouTube or social apps, and no live audio.
• Subtitles come as SRT/VTT files rather than burned into the picture.

Captions in the spoken language, or subtitles in another

There are two different jobs here. Same-language captions serve viewers who watch muted or can't hear the audio; run transcript mode with the spoken language selected and the SRT and VTT files come back untranslated, in the language spoken. Translated subtitles serve viewers who don't understand the speech; pick a different target language and the files come out in that language instead.

For accessibility, captions usually do more than repeat the words: deaf and hard-of-hearing viewers rely on cues such as [applause] or speaker names. These files contain spoken words only, so add those by hand where they matter. Captioning sound effects and music covers the conventions.

Transcript-mode or dub-mode subtitle files

Transcript mode, spoken language selected

Captions in the spoken language, untranslated, 1 credit per minute

Transcript mode with a target language

The full translation of each segment, 1 credit per minute

Dub mode

Subtitles that follow what the translated voice says, including lines shortened to fit timing; included in the 50 credits per minute

If you publish a dubbed version and also offer subtitles, use the dub-mode file so text and voice agree. If you publish subtitles over the original audio, the transcript-mode file is the fuller translation.

A sample cue, and what to check before publishing

• Reading time: a long sentence spoken quickly can flash past. Split it into two cues in a subtitle editor.
• Line length: players wrap long lines differently, so preview on the device your audience uses.
• Names and terms: correct them once in the file rather than in every platform's caption editor.
• Right-to-left languages: Arabic and Hebrew subtitles need a player that handles right-to-left text; check punctuation at line ends. The Arabic page covers this.
• Odd characters: if accents turn into symbols, the player is reading the file with the wrong text encoding.

Editing SRT files covers fixing text and shifting timing, and splitting text into cues covers where line breaks should fall.

Generate subtitles from the final cut

Subtitle timings are tied to the exact file they were made from. If you trim the intro, move a scene or add a title card after generating them, every cue after the change drifts. The fix is ordering: lock the edit, export, then run that export. If you deliver several versions, such as a full episode and a short cut for social, run each export separately rather than trying to re-time one file for both.

For long projects edited in stages, a useful pattern is to generate a quick draft from the rough cut for reviewers to read, then a clean run on the final export for publishing. Before release, a short subtitle quality check catches timing, spelling and reading-speed problems in one pass.

Building a set of subtitle languages

Each run produces one subtitle language. A 30-minute episode with English captions plus Spanish, French and German subtitles is four runs of 30 credits: 120 credits (12¢). Name the files consistently, such as episode.en.srt and episode.es.srt, which many desktop players pick up automatically when the file sits next to the video. On YouTube, each file is uploaded as its own language; the YouTube subtitles use case and the guide to translating a YouTube video cover that route. Showing two languages at once is possible on some players; see dual-language subtitles.

When a sidecar file isn't what you need

Some social apps and ad platforms don't accept subtitle files, and others display them inconsistently. In those cases the text has to be part of the picture, which mydubly doesn't do; import the SRT into an editor and export a captioned version from there. Embedded, sidecar or burned-in subtitles compares where each works. Subtitles also arrive after the recording, never during it: there are no captions for live streams.

If you only need the text for reading, the timestamped transcript is easier to work with than a subtitle file.