Free Audio & Video Transcriber
Turn any audio or video into text and download ready-to-use SRT/VTT captions — powered by Whisper running entirely on your device. No upload, no API key.
How it works
Pick a model & file
Choose English or Multilingual, then select any audio or video file. It's decoded locally.
Whisper runs on-device
The model downloads once (cached after), then transcribes with timestamps — nothing leaves your device.
Copy or export captions
Copy the transcript, or download it as SRT, VTT, or TXT for YouTube and editors.
Need the full studio, not just a quick fix?
PandaStudio records, transcribes, captions, edits, and exports — driven by AI agents or a real timeline, all running locally on your Mac or PC. This free tool is one piece of it.
Frequently asked questions
Is my audio or video uploaded to transcribe it?+
No. The tool runs OpenAI's Whisper model compiled to WebAssembly right in your browser. Your file is decoded and transcribed on your own device — nothing is uploaded and no API key is needed.
Can I get subtitles / captions?+
Yes. Along with plain text, you can download timed captions as .srt or .vtt, ready to drop into YouTube, a video editor, or PandaStudio.
Does it work for languages other than English?+
Yes — pick the Multilingual model before choosing your file. The English model is faster; the multilingual one handles many languages.
Why is the first run slow?+
The Whisper model weights download once (tens of MB) and are cached by your browser. After that, transcription speed depends on your device and the length of the file.
How accurate is it, and what about long files?+
It uses the tiny/base Whisper models for speed in-browser, which are good for most clips. For long recordings, higher accuracy, speaker labels, and editing the transcript directly against the video, use the PandaStudio desktop app.