Subtitle Generator
Generate SRT subtitles from audio or video using AI speech recognition.
Drop an audio or video file here, or click to browse
Accepted: audio/*,video/*
Click or drag to upload
Browser-Based
Runs entirely in your browser. No data leaves your device.
AI-Powered
Uses OpenAI Whisper. Best on clear speech — podcasts, interviews, lectures.
SRT Export
Download subtitles in standard SRT format.
This free subtitle generator turns speech into timed SRT captions automatically using OpenAI’s Whisper speech-recognition model. Drop in an audio or video file — a podcast episode, interview, lecture, or screen recording — and get back a standard .srt subtitle file you can upload to YouTube, import into Premiere or DaVinci Resolve, or play alongside your video in VLC. There is no sign-up and no watermark, and unlike most transcription services the AI runs entirely in your browser, so your recording never leaves your device.
You can choose between two Whisper models: Tiny (~75 MB) for the fastest results, or Base (~145 MB) when accuracy matters more. The model downloads once on first use and is cached for every session after that. Language can be auto-detected or set explicitly from a dozen options including English, Spanish, French, German, Chinese, Japanese, and Arabic. Whisper performs best on clear spoken word — podcasts, interviews, lectures, and voiceovers — and the tool cleans up timing, fills gaps, and flags low-confidence results before you download.
How to generate SRT subtitles from a video
- 1Drag and drop an audio or video file into the upload zone, or click it to browse — files up to 500 MB are accepted.
- 2Pick a Model: Tiny (~75 MB) is the fastest, while Base (~145 MB) produces noticeably better transcriptions.
- 3Choose the spoken Language, or leave it on Auto-detect and let Whisper work it out.
- 4Click Generate Subtitles. On first use the model downloads and is cached; the audio is then decoded and transcribed right in your browser.
- 5Review the timed segments in the on-screen preview, then click Download SRT to save the subtitle file.
Frequently asked questions
What file types can I generate subtitles from?
Any audio or video file your browser can decode, which covers the common cases: MP4, WebM, and MOV video, plus MP3, WAV, M4A, and OGG audio. The tool extracts and resamples the audio track automatically before transcribing.
How accurate are the generated subtitles?
Whisper is remarkably good on clear spoken word — podcasts, interviews, lectures, and voiceovers — and the Base model is noticeably more accurate than Tiny. It struggles with music, singing, heavy background noise, and overlapping speakers, and the tool will warn you when a result looks unreliable. Treat the output as a strong first draft and give it a quick proofread before publishing.
Which languages are supported?
You can auto-detect the language or explicitly select English, Spanish, French, German, Italian, Portuguese, Chinese, Japanese, Korean, Hindi, Arabic, or Russian. Whisper was trained on multilingual audio, so accuracy is strong across all of them, with English being the strongest.
Is my audio uploaded to your servers?
No — this tool is unusual in that transcription happens entirely in your browser. Your file is decoded and processed locally by the Whisper model running on your own device; the recording itself is never uploaded anywhere. The only network traffic is the one-time download of the AI model, which is then cached for future visits.
Why does the first run take longer?
The first time you generate subtitles, your browser downloads the Whisper model — about 75 MB for Tiny or 145 MB for Base. It is cached after that, so later runs skip straight to transcription. Long recordings also simply take longer to transcribe, since the work happens on your own hardware.
Is this subtitle generator free?
Yes — completely free, with no sign-up, no watermarks, and no per-minute transcription charges, because the processing runs on your own device.
Enjoying this tool?
Toolism is free and always will be. If it saved you time, consider buying me a coffee!
Buy me a coffee