Skip to content
TabShed
My storage

Subtitle Generator – Make SRT & VTT Captions

Generate subtitles from any video or audio with AI. Edit the timing, download SRT or VTT. Free, no watermark, nothing is uploaded.

Processed on your device · no upload

Drop a video or audio file to make subtitles

MP3, WAV, M4A, MP4, MOV, MKV, WebM and more · up to 500 MB · or paste with Ctrl+V

Runs on your device · nothing is uploaded. The first run downloads a speech model once (41–249 MB, depending on the quality you choose).

What to do next

Audio to TextJust need the transcript as text or Word?Video to AudioExtract the audio from a videoVideo CompressorShrink the video before you upload it with its subtitlesAudio CutterTrim an audio file before transcribing

How to use Subtitle Generator

  1. Add your video or audio

    Drop an MP4, MOV, MKV, WebM, MP3 or WAV file (up to 500 MB). You can preview it with the subtitles right on the page.

  2. Choose language and cue length

    Pick the spoken language (or leave it on automatic) and how long each subtitle may be. Turn on Translate to English for foreign-language videos.

  3. Generate

    The first run downloads the speech model once. Subtitles are timed to the speech and you can cancel at any time.

  4. Edit and download

    Fix any word or time in the list, click a cue number to jump there in the player, then download SRT (YouTube, Premiere, VLC) or VTT (web players).

Frequently asked questions

Is there a watermark or a limit?
No watermark, no sign-up and no minute limit. The only limits are your device: files up to 500 MB, and very long videos take a while without a GPU.
Does it burn subtitles into the video?
No. It gives you an SRT or VTT file that you upload next to the video (YouTube, Vimeo, Facebook, LinkedIn) or load in a player such as VLC. Burning captions into the picture is not part of this tool.
Is my video uploaded?
No. The audio is read and recognised in your browser; the video never leaves your device. The speech model downloads once and is cached.
Which AI model does this use?
OpenAI Whisper (tiny, base, small), MIT licence, as ONNX files by the Hugging Face community (onnx-community/whisper-*), run with Transformers.js (Apache-2.0) and ONNX Runtime Web (MIT), loaded from the Hugging Face Hub.
How exact is the timing?
Each subtitle starts within about a second of the speech. Long sentences are split into short cues and the time is shared by length, so fine-tune start and end times in the list when you need frame-exact timing.
What does Translate to English do?
Whisper can write English subtitles for speech in other languages. It only translates into English, and the small models make mistakes, so check the text.

More video tools

Blur Faces in VideoTrack and blur faces in a videoVideo to AudioExtract the soundtrack as MP3 or WAVVideo CompressorShrink videos to a target size