Skip to main content

Generate Subtitles from Video & Audio: Free

Auto-generate subtitles from any video or audio file with AI speech recognition.

Rated 4.5 out of 5 from 79 ratings

By uploading, you confirm you have the rights to these files and agree to our Terms.

Encrypted in transit

How to Convert

  1. 1

    Upload your video or audio file by dragging it into the upload area or clicking to browse.

  2. 2

    Choose your output format: SRT subtitles, VTT captions, or a plain-text transcript.

  3. 3

    Click Generate: our AI transcribes the audio track and builds timestamped subtitles automatically.

  4. 4

    Download your subtitle file, or edit it first: SRT and VTT are plain text and open in any editor.

What you get

  • AI speech recognition with word-level timestamps
  • SRT, VTT, or plain-text transcript output
  • Auto-detects the spoken language (~50 languages)
  • No watermark, no signup required
  • Guest files auto-delete in about 2 hours
Supported Formats
.MP4, .MOV, .MKV, .WEBM, .AVI, .MP3, .WAV, .M4A, .AAC, .FLAC, .OGG

Limits by plan

No account
Files up to 1 GB
10 conversions a day
Free account
Files up to 1 GB
20 processing minutes a day
Videos up to 30 minutes long (1080p)
Pro
Files up to 2 GB
1,500 processing minutes a month
20 files at once
Business
Files up to 20 GB
4,000 processing minutes a month
50 files at once

Compare the plans

Frequently Asked Questions

Very good on clear speech: a single speaker, a decent microphone, and minimal background noise typically produce near-publication-ready subtitles that need only a quick proofread. Accuracy drops with heavy accents, multiple people talking over each other, loud background music, or low-quality recordings: expect to spend more time editing in those cases.
The spoken language is detected automatically: there is nothing to configure. Around 50 languages are supported, including English, Spanish, French, German, Portuguese, Italian, Japanese, Korean, and Chinese. Recognition quality is strongest for widely spoken languages.
Free accounts can upload files up to 1 GB and 30 minutes of audio. Pro users can transcribe recordings up to 2 hours long. If your video is too large, compress it first or extract just the audio track: transcription quality is identical either way.
Yes. Files are transferred over HTTPS, processed in an isolated server-side directory, and automatically deleted (after about 2 hours without an account; account files follow your plan). You can also delete them immediately from the download card. We never use your audio to train models or share it with third parties.
Yes. SRT and VTT are plain-text formats: open them in any text editor to fix a word or adjust a timestamp. Dedicated subtitle editors like Subtitle Edit or Aegisub add a waveform view for finer timing tweaks.
Yes. Set "Translate Subtitles To" before generating and the transcript is machine-translated line by line into your chosen language (15 available, including Spanish, French, German, Japanese, and Chinese) while keeping every timestamp. Like all machine translation it's solid for comprehension; give it a quick proofread before publishing.
Video: MP4, MOV, MKV, WebM, and AVI. Audio: MP3, WAV, M4A, AAC, FLAC, and OGG. For video uploads we extract the audio track server-side before transcribing, so the video codec does not matter.
No, it produces a sidecar subtitle file (SRT or VTT) that you load alongside the video; players like VLC and platforms like YouTube pick it up automatically. To permanently burn subtitles into the video frames, upload your video and the SRT to our video converter and use its subtitle option.
Learn more about Subtitle Generator

About Subtitle Generator

Every video published without subtitles leaves viewers behind. Most social feeds autoplay muted, search engines cannot watch a video but can read a transcript, and a meaningful share of your audience relies on captions to follow along at all. Our subtitle generator turns any video or audio file into accurate, timestamped subtitles in minutes: no typing, no watermark, no signup required.

How it works. Upload a video (MP4, MOV, MKV, WebM, AVI) or an audio file (MP3, WAV, M4A, AAC, FLAC, OGG). Our servers extract the audio track and run it through an AI speech recognition model that transcribes with word-level timestamps. Those timestamps are then segmented into properly paced caption lines: no walls of text, no captions flashing past too fast to read. Download the result as an SRT file (the universal SubRip standard accepted by YouTube, VLC, and every editing suite), a VTT file (the HTML5 WebVTT format for web players), or a plain-text transcript with no timestamps at all.

Language is detected automatically. There is no dropdown to fiddle with: the model identifies the spoken language across roughly 50 languages and transcribes in that language. Widely spoken languages like English, Spanish, French, German, Portuguese, and Japanese get the strongest accuracy: clear speech recorded on a decent microphone routinely comes back near publication-ready, while heavy accents, people talking over each other, and loud background music will need more editing afterwards. Since SRT is plain text, fixing a word takes seconds in any editor.

Free and private. The free tier covers files up to 30 minutes and 1 GB (enough for most YouTube videos, lectures, and podcast segments) while Pro extends the duration cap to 2 hours. Like every tool on this site, files are transferred over HTTPS and auto-delete (after about 2 hours without an account; account files follow your plan), with a "Delete file now" button on the download card if you would rather not wait.

The subtitle file you get back is a sidecar: it sits next to the video rather than inside it. To permanently burn the captions into the picture, run the video and SRT through our video converter and use its subtitle option. If you only need the soundtrack, extract the audio first to shrink the upload. And for meeting recordings that need speaker labels and AI summaries, our dedicated transcription product Convert Audio to Text is built exactly for that.

Common Use Cases

  • Add captions to YouTube, TikTok, and Instagram videos to boost watch time and accessibility
  • Generate SRT files for course videos and training material without typing a single timestamp
  • Transcribe interviews, lectures, and podcast episodes into searchable text
  • Meet accessibility requirements with accurate closed captions for every published video
  • Create subtitle files for client video deliverables in minutes instead of hours

How It Works

Subtitles are generated server-side by an automatic speech recognition model. The audio track is extracted from your upload, transcribed with word-level timestamps, and segmented into readable caption lines with standard line-length and duration pacing. The spoken language is auto-detected across roughly 50 languages. Output is available as SRT (SubRip), VTT (WebVTT), or a plain-text transcript, and, without an account, source files auto-delete after about 2 hours.