Skip to content
DYNAMB

Auto Subtitle Generator — Free & Private

Generate subtitles automatically with this free auto subtitle generator — OpenAI's Whisper model runs inside your browser, so your video and audio never upload anywhere. Drop a recording, approve the one-time model download, and export ready-to-use SRT or VTT files with correct timestamps, in 90+ languages including Hindi.

  • No upload — files stay on your device
  • Works offline once loaded
  • Free forever, no signup, no watermark
Drop an audio or video file, or tap to choose MP3, WAV, M4A, OGG, MP4, WebM… — the recording never leaves your device
{{ decodeStatus || 'Decoding audio…' }}
{{ file.name }} {{ fileSizeLabel }} · {{ durationLabel }} of audio
{{ slowBanner }}
AI model size

Your recording is processed entirely on this device — it is never uploaded. The only download is the open-source Whisper model itself, fetched once from Hugging Face's servers and then stored in your browser cache for next time.

{{ modelStatus }} ({{ downloadLabel }})

One-time download — the model is cached after this, and your audio still never leaves the device.

Transcribing on {{ device === 'webgpu' ? 'your GPU (WebGPU)' : 'your CPU (WASM)' }} — {{ progressPct }}%
  • {{ timeLabel(seg.start) }} {{ seg.text }}
  • …{{ livePartial }}

Done — {{ segments.length }} segments transcribed on your device.

Stopped early — keeping the {{ segments.length }} segments finished so far.

{{ qualityWarning }}

  • {{ timeLabel(seg.start) }} {{ seg.text }}

How to use this tool

  1. Drop a video or audio file

    MP4, WebM, MKV, MP3, WAV and more — the audio track is extracted and decoded entirely on your device.

  2. Choose a model size and download it once

    Tiny (~75 MB) is fast but rougher; Base (~145 MB) is the recommended balance. The model is cached after the first download, so later visits start straight away.

  3. Export SRT or VTT

    Watch the subtitles stream in live, then download an .srt for editors and players or a .vtt for the web — long lines are split automatically to subtitle-friendly lengths.

Questions people ask

How do I generate subtitles automatically for free?

Drop your video above — Whisper AI transcribes it on your device and exports a ready-to-use SRT or VTT file, free with no upload, no watermark and no sign-up.

What is the difference between SRT and VTT?

Same subtitles, different dialects: SRT numbers each cue and writes timestamps with a comma (00:01:02,500), while VTT starts with a WEBVTT header and uses a dot (00:01:02.500). Editors and desktop players usually want SRT; HTML5 video and most web players want VTT. This tool exports both.

Is my video uploaded anywhere?

No. The audio track is decoded and transcribed entirely inside your browser — this site has no upload endpoint for it. The only download is the open-source AI model itself, once, from the Hugging Face CDN, and only after you click to approve it.

Which model size should I choose?

Tiny (~75 MB) for quick drafts and slower devices, Base (~145 MB) when you care about accuracy — it makes noticeably fewer mistakes and is what we recommend for subtitles you intend to publish.

Can it subtitle Hindi or mixed-language videos?

Yes — Whisper supports Hindi and 90+ other languages. Auto-detect handles most files; choosing the language explicitly gives better results on accented or code-switched speech like Hinglish.

How long does a video take to subtitle?

On a desktop with WebGPU, a few minutes of video takes well under a minute. Without GPU acceleration the tool falls back to a slower CPU path — it still works, but a long video takes a while, so the page shows honest per-chunk progress and lets you stop early and keep what is done.

Related tools

Free because it costs us nothing to run: your browser does the work, not our servers. Made by Dynamb Technologies — we build software for businesses.