Speech to Text

Turn recordings and videos into text and subtitles (SRT) — privately, with AI.

Choose a recording or video

MP3, M4A, WAV, Ogg, voice notes, MP4, MOV, WebM · up to 60 minutes · stays on your device

or drop it here

Processed in your browser. Your files are not uploaded.

Share this app

Know someone who needs this? Send them straight to it — the link opens this exact tool.

Download the share imageFor Instagram, TikTok or a status, where links do not show a preview.

The preview card this link shows when shared
How this link appears on Facebook, WhatsApp and X.

About this tool

Transcribe voice notes, interviews, lectures, meetings and videos into editable text, and download subtitles as SRT or VTT for YouTube, Facebook, Reels and TikTok. It runs OpenAI's Whisper speech-recognition model in your browser, so recordings are never uploaded; the model (about 80 MB) is downloaded from this site the first time and then kept by your browser. Recognises English and 90+ other languages including Nepali and Hindi, and can translate speech into English. Accuracy is best for clear English; check names and numbers.

How to use Speech to Text

  1. Choose a recording or a video — a voice note, interview, lecture, meeting or reel.
  2. Pick the language spoken (or let it detect), and tick Translate into English if you want English text.
  3. Press Transcribe. The first time, the speech model is downloaded from this site; after that it starts straight away.
  4. Correct anything it misheard, then copy the text or download it as .txt, or as .srt / .vtt subtitles.

Questions

Is my recording uploaded?

No. The audio is decoded and transcribed in your browser by OpenAI's Whisper model. Only the model itself is downloaded, from this site, the first time.

How accurate is it?

Very good for clear English speech; good for most major languages. Noise, music, several people talking at once and strong accents reduce accuracy. Nepali and Hindi work, but expect more corrections. Always check names and numbers.

How do I add the subtitles to a video?

Download the .srt file. YouTube: Subtitles → Add language → Upload file. Facebook: Edit video → Subtitles & captions → Upload SRT. CapCut and Premiere import .srt too. Use .vtt for web players.

How long can the recording be?

Up to 60 minutes. It takes roughly as long as the recording on a laptop, longer on a phone — keep the tab open. For longer recordings, split them with Audio Cutter.

Is this real AI?

Yes — Whisper is a neural speech-recognition model trained by OpenAI and released under the MIT licence. This app runs the 'base' version through transformers.js and ONNX Runtime.

More free apps

All apps →

Do this at scale with AI

This app handles the one-off job. These tools handle it as part of a real workflow.