Auto Subtitle Generator: Generate SRT subtitles from any video automatically, privately. Input: Audio, video, MP3, M4A, WAV, OGG, FLAC, AAC, MP4, WEBM, MOV, MKV files and values you enter. Output: Results shown on the page, a downloaded file and text to copy. Privacy: Runs in your browser: inputs and files stay on your device and are not uploaded. On first use it downloads an open AI model to your browser; processing still happens locally.
How to generate subtitles automatically
- Drop your video (MP4, MOV, WebM) or its audio onto the box. Only the sound is used, and it stays on your device.
- Pick the accuracy and the language spoken. Choose Translate to English to make English subtitles for a video in another language.
- Press Transcribe. The first time, the Whisper speech model downloads once; the size is shown before you start.
- Click lines to check them against the audio, then press Download .srt (or .vtt).
- Upload the SRT to YouTube, Vimeo, LinkedIn or Facebook, or import it into Premiere Pro, DaVinci Resolve, Final Cut or CapCut. Use the subtitle editor to fix words or timing.
Why add subtitles?
Most social video is watched with the sound off, and captions keep viewers watching, make videos accessible to deaf and hard-of-hearing people, help non-native speakers and give search engines text to index. Typing them by hand takes hours. This generator uses OpenAI's open Whisper model, running in your browser, to recognise the speech and create timed captions split into readable two-line cues, aiming for up to 42 characters per line.
Because everything runs locally there are no per-minute charges, watermarks or upload limits, and unreleased or confidential videos stay private. Automatic captions are a fast first draft: names, brand terms and numbers may need correcting, and speakers are not labelled. Clear audio with little background music gives the best results.
Example: three 4-minute videos
We generated subtitles for three 4-minute English readings from LibriSpeech, a standard public speech test set, with the default Balanced (Whisper base) model in Chrome on a laptop. Each took about 30 seconds. The word error rate was 3.0%, about 3 words wrong in 100, against 3.9% for the same model in the desktop program whisper.cpp. Half of the subtitle start times were within 0.16 to 0.36 seconds of the moment the words were spoken, depending on the video. None of the 207 cues was longer than 84 characters and 92% kept every line within 42 characters; whisper.cpp's default SRT met that line length in about a third of its cues.
How it compares with free online subtitle makers
| This tool | Kapwing (free) | VEED (free) | |
|---|---|---|---|
| Auto-subtitle minutes | No limit | Up to 10 minutes | Not stated |
| Watermark | None (you download an SRT or VTT file) | Watermark on exports | Not stated |
| Video export | Not needed: add the SRT on your platform, or burn it in | 720p, up to 4 minutes | Not stated |
Free limits as published by each service (checked 22 October 2026): Kapwing pricing, VEED pricing. They change often; the paid plans lift them.
FAQ
Which subtitle formats can I download?
SRT, which almost every platform and editor accepts, and WebVTT for web players. Plain text is also available.
Is my video uploaded?
No. The audio is extracted and transcribed in your browser; only the speech model is downloaded.
Can it subtitle videos in other languages?
Yes. Whisper recognises about 100 languages and can also translate the speech into English subtitles.
How accurate are the captions?
Very good for clear speech in major languages, but always review names, jargon and numbers before publishing.
How do I add subtitles to a video for free without a watermark?
Generate the SRT here and upload it next to your video on YouTube, Vimeo, LinkedIn or Facebook, which show it as captions. To put the words into the picture itself, use Burn subtitles into video. Neither adds a watermark.