Skip to content

Speech to text · Auto-detect + 10 languages · burned in · up to 4 min

Auto subtitle generator — speech in, burned-in subtitles out

Upload a video and the AI transcribes the speech, times every line and burns the subtitles into the picture. Below: how the language setting works, what burned-in means next to an SRT file, and what to use when you need a file.

  • Speech transcribed automatically
  • Auto-detect or pin the language
  • Burned into one MP4
  • Original vs Captioned preview

Updated by the

Quick answer

Updated

What is an auto subtitle generator?

An auto subtitle generator uses AI speech recognition to turn the dialogue in a video into timed subtitles without any typing. Kenerate's version detects the spoken language automatically, or lets you pin one of 10, and burns the subtitles into the video as one MP4 at 720p or 1080p. It doesn't export SRT or VTT files.

Key facts

Spoken languages
Auto-detect + 10
Subtitle type
Burned in (open)
SRT / VTT export
No
Translation
No — dub first
Clip limit
4 min, 100 MB
Formats in
MP4, MOV, WebM

Real demo

Speech in, subtitles out

Drag the line. Left: the classroom clip as recorded. Right: the auto subtitle generator's output in Plain — transcribed and timed with no typing.

OriginalSubtitled
  • Original. The 16:9 clip with speech, no subtitles.
  • Subtitled. Plain style at the bottom, burned into the frame.

Both are real, unedited files, shown at the same size.

Subtitle a clip

Real results

Four clips, subtitled automatically

Real Wan 3 talking clips with English speech. Nobody typed a word: each was transcribed and subtitled by the tool, in a different style.

Clips: Wan 3 image to video with speech (AI-generated people). Subtitles: Kenerate Auto Subtitles, unedited.

How to

How to generate subtitles automatically

Time needed
Steps
4 steps
  1. Step 1

    Check the audio

    One clear voice at a time and low music give the best transcript.

  2. Step 2

    Upload the video

    MP4, MOV or WebM, up to 4 minutes and 100 MB; split longer videos into parts.

  3. Step 3

    Set the spoken language

    Leave Auto-detect, or pin one of the 10 languages the tool lists.

  4. Step 4

    Pick style, position and resolution

    10 styles; Top, Middle or Bottom; 720p or 1080p HD.

Kenerate AI Auto Subtitles with a talking-head clip loaded as the video to caption

Open Auto Subtitles and upload a clip, or pick one from your history, library or a link.

1 / 5 · Add your video

Watch the tour · 1:19

Text, image & reference to video, models, edit, extend & chat

Chapters

Engines

Which spoken-language setting to use

The tool's language list, and when each option makes sense.

  • Auto-detect title card — a glowing globe made of soft sound waves floating above a podcast microphone in a dark studio
    Default

    Default setting

    Auto-detect

    Input
    Any clip up to 4 min
    Output
    Language picked from the audio

    One clear language, everyday accents

    Open Auto-detect
  • English (US) title card — a fictional adult creator talking to a phone on a rooftop with an American city skyline at sunset behind

    Spoken language

    English (US)

    Input
    Any clip up to 4 min
    Output
    English subtitles

    American and most other English

    Open English (US)
  • English (India) title card — a fictional adult creator talking to a phone on a lively Indian city street at golden hour

    Spoken language

    English (India)

    Input
    Any clip up to 4 min
    Output
    English subtitles

    Indian-accented English

    Open English (India)
  • Spanish · French · German · Italian · Portuguese title card — a fictional adult creator talking to a phone in a sunny old European plaza with a fountain and café tables

    Spoken language

    Spanish · French · German · Italian · Portuguese

    Input
    Any clip up to 4 min
    Output
    Subtitles in that language

    European-language speech; pin it for accents

    Open Spanish · French · German · Italian · Portuguese
  • Japanese · Chinese title card — a fictional adult creator talking to a phone on a rainy East Asian city street at night

    Spoken language

    Japanese · Chinese

    Input
    Any clip up to 4 min
    Output
    Subtitles in that language

    Japanese or Chinese speech

    Open Japanese · Chinese
  • Russian title card — a fictional adult creator in a warm coat talking to a phone on a snowy Moscow-style square with onion domes softly blurred behind

    Spoken language

    Russian

    Input
    Any clip up to 4 min
    Output
    Russian subtitles

    Russian speech

    Open Russian
All models (7)
  • Hindi and other languages title card — a dubbing booth with a studio microphone

    Not in Auto Subtitles

    Hindi and other languages

    Input
    Speech in an unlisted language
    Output
    Dub into a listed language first

    AI dubbing (33 target languages), then subtitle

    Open Hindi and other languages

Subtitles come out in the language that is spoken; the tool doesn't translate.

Tips

Getting a clean auto transcript

The audio decides most of the result.

  • Pin the language when a speaker has a strong accent.

  • Choose English (India) for Indian-accented English rather than English (US).

  • Keep background music well below the voice.

  • Cut long intros with no speech before uploading.

  • Split a long video into parts under 4 minutes each.

  • If speakers switch languages, pin the main one.

Example uses

How creators use Auto subtitle generator

Illustrative examples of typical workflows, not customer reviews.

  • Language teacher speaking to camera at a classroom desk

    Language teacher

    A Spanish teacher can subtitle short speaking clips in Spanish so learners can read along while they listen.

  • Volunteer speaking to camera in a community garden

    Nonprofit team

    A nonprofit can subtitle a volunteer's short interview once and share the same file on its website and social pages.

  • Retired watchmaker speaking at his workbench

    Documentary maker

    A filmmaker can burn subtitles into a festival teaser where the speaker's accent is hard to follow.

FAQ

Auto Subtitle Generator — questions

How does an auto subtitle generator work?

It runs speech recognition on the video's audio, splits the transcript into short lines with start and end times, and shows them in sync with the voice. Kenerate's version then burns those lines into the frames in your chosen style, so you download one finished MP4 instead of a separate subtitle file. A run takes about 1–3 minutes.

Link to this answer

Can I burn subtitles into a video?

Yes — burning subtitles into the video is the only output here. Kenerate's auto subtitle generator draws the subtitles into every frame and returns one MP4 at 720p or 1080p, so they show in any player or app without a separate file. The trade-off: viewers can't turn them off, and the text can't be changed after export.

Link to this answer

Can I download an SRT or VTT file?

No. Kenerate burns subtitles into the video and doesn't export SRT, VTT or any other subtitle file. If you need a file, for a YouTube subtitle upload, a website player or a translator, use a tool that exports one: OpusClip lists TXT, SRT and VTT, and VEED lists SRT, VTT and TXT on their own pages, checked 3 October 2026.

Link to this answer

Which languages does the AI subtitle generator support?

Auto-detect plus 10 spoken-language options: English (US), English (India), Spanish, French, German, Japanese, Chinese, Italian, Portuguese and Russian. Hindi isn't available yet. Subtitles always come out in the spoken language, because the tool transcribes and doesn't translate. For other languages, dub the video with Kenerate's AI dubbing first, then subtitle the dubbed version.

Link to this answer

How do I get subtitles in another language?

Dub first, then subtitle. Kenerate's AI dubbing translates and re-voices a video into one of 33 target languages, keeping each speaker's voice. Run the dubbed video through Auto Subtitles with the new language pinned, and you get subtitles in that language. This works when the dub language is one the subtitle tool lists, such as Spanish, French, German or Japanese.

Link to this answer

Is the subtitle generator accurate with accents?

It is most accurate on clear speech in a listed language. For accented English, pin English (US) or English (India) rather than relying on Auto-detect; for other languages, pin the language you hear. Background music, crosstalk and echo cause most errors, and because subtitles can't be edited after generation, cleaning up the audio before uploading matters more here than in editable tools.

Link to this answer

How long can the video be?

Up to 4 minutes and 100 MB per video, in MP4, MOV or WebM. For a longer video, cut it into parts under 4 minutes, subtitle each part and join them in your editor. Pricing goes by length tier, up to 1, 2, 3 or 4 minutes, with no subscription required, and new accounts get starter credits to try it.

Link to this answer

Open vs closed subtitles

Open subtitles are drawn into the video frames and can't be removed; closed subtitles are stored separately and overlaid by the player, so viewers can switch them on or off or choose another language [1]. Kenerate's auto subtitle generator makes open, burned-in subtitles only.

Explore

Every Kenerate video tool

Generators, tools, models and guides — all in one studio.

See more of Kenerate AI in your Google resultsAdd kenerateai.com as a preferred source — Google shows more from your preferred sources in Top Stories and AI answers.

Subtitle a video now

One studio for images and video.

Tried it? Tell us how it went