Rescroll

Give ChatGPT a video as a transcript it can read.

ChatGPT's file uploads take documents, images and audio, not video. Drop the video here: it's transcribed on your device, every line timestamped, ready to paste into ChatGPT with your question. Free, and nothing is uploaded.

  • 100% free
  • No sign-up
  • 99 languages
  • Timestamps, SRT and VTT
  • Files stay on your device

Loading the tool…

It runs in your browser. Nothing is uploaded.

How it works

How to convert video to text, free, in three steps.

  1. 1

    Drop in your video or audio

    MP4, MOV, WebM, MP3, M4A or WAV, from a phone, a camera, a screen recorder or a podcast mic. It opens in your browser; nothing is uploaded.

  2. 2

    Transcribe it

    The language is worked out from the first thing said, or pick it yourself. The first time, the speech model downloads once and stays in your browser for next time.

  3. 3

    Edit, copy or download

    Read the transcript line by line with timestamps, fix any word, then copy it or download it as plain text, SRT or WebVTT captions.

Why this one

Free, with nothing held back, and nothing uploaded.

Every word timed

Each line shows when it starts; click the time and the video plays from there. Captions are timed to the word, not spread evenly over a sentence.

Captions ready to upload

SRT and VTT files cut the way subtitle editors cut them: two lines of up to 42 characters for widescreen video, one short line at a time for vertical video.

99 languages, detected for you

It runs Whisper, OpenAI's open speech-recognition model, which knows 99 languages. It works out which one is spoken, or you choose it.

Private by design

The model runs on your own graphics card or processor, in the browser. Your video is never uploaded, so it's fine for unreleased videos, client calls and anything else you'd rather keep to yourself.

Your file is opened and processed by your own browser. It is never sent to a server, so it stays private, and there’s no upload to wait for.

Why ChatGPT needs the transcript, not the video

OpenAI's help center lists the files ChatGPT works with: documents, spreadsheets, presentations, images and audio. Video isn't on the list, and its audio uploads turn away MP4 and WebM files that contain video. Audio itself needs a paid plan, and GPT-6 at Instant takes recordings of up to six minutes.

Text works on every plan. This page turns the video's speech into a transcript on your device, every line with the time it's said, so when you ask what the speaker says about pricing, ChatGPT can tell you where. Paste it into the chat, or download it as a TXT file and attach it: ChatGPT reads text files of up to 2 million tokens.

What a transcript can't carry is the picture: text on screen, slides and what's shown aren't in the speech. Rescroll reads both from a TikTok, Reel or YouTube link, and lets you chat with the video itself.

Rescroll

Only have the link? Need the words on screen too?

This converter transcribes a file you have. Rescroll goes further: paste a TikTok, Instagram Reel or YouTube link and it gets the transcript and reads the text on screen, then lets you ask anything about the video.

  • Transcripts straight from a link
  • On-screen text read too
  • Chat with every video you save

Questions

The video to text converter: your questions, answered.

Can I upload a video to ChatGPT?

Video isn't on OpenAI's list of files ChatGPT works with, and its audio uploads turn away MP4 and WebM files that contain video. The dependable way in is the transcript: make it here, copy it, and paste it into the chat with your question.

How do I get ChatGPT to summarize a video?

Transcribe the video on this page, copy the transcript and paste it into ChatGPT with what you want: a summary, the key points, quotes with their times. For a long video, download the TXT and attach it instead of pasting.

Why are timestamps on?

So ChatGPT can say when something is said, not only what: each line starts with its time, and answers can point you to the moment. For plain text, switch off Timestamps in the text, under Take it away, before you copy.

Can't I just send ChatGPT the video's sound?

On a paid plan, yes: ChatGPT takes audio files of up to 512 MB, and GPT-6 at Instant transcribes recordings of up to six minutes. For longer videos, or on the Free plan, the transcript is the way in. To get a video's sound as a file, use the MP4 to MP3 converter.

How accurate is the transcription?

It uses Whisper, OpenAI's open speech-recognition model: the base version by default, and the larger small version when you choose High accuracy. On clear speech in widely spoken languages it gets nearly every word; strong accents, people talking over each other, music under the voice and unusual names are where it slips. Read it through and fix what's wrong: the captions update with your edits.

Which languages can it transcribe?

99, from English, Spanish, Portuguese, Hindi and Arabic to Welsh and Swahili: every language Whisper was trained on. It detects the language from the first speech in the file; if it guesses wrong, pick the language and transcribe again. It's most accurate in widely spoken languages; for others, choose High accuracy.

Does it read the text on screen?

No. It transcribes what's said. Text that only appears on screen, like a hook in big letters, captions burned into a TikTok or a slide, isn't in the sound, so a speech model can't hear it. Rescroll reads both: paste a TikTok, Reel or YouTube link and it transcribes the speech and reads the on-screen text, frame by frame.

Why does it download a model the first time?

Because the transcription happens on your device, the speech model has to be on your device too. The first time, your browser downloads it from Hugging Face, usually 110 to 140 MB with the speech engine (about 280 to 350 MB for High accuracy; the page shows the exact size for your device first), and keeps it, so the next file starts in seconds. Only the model is downloaded: your video isn't sent anywhere.

Which browsers does it work in?

Current Chrome, Edge, Safari and Firefox. In Chrome and Edge on a computer the speech model runs on your graphics card (WebGPU), the fast way; where a browser can't do that, it runs on your processor instead and takes longer. On a recent laptop, with the model already downloaded, a 37-second clip took about 4 seconds on the graphics card and about 10 on the processor.

How big a file can it handle?

There's no file-size limit of ours: a video of several gigabytes is fine, because only its sound is kept, and long recordings are transcribed in pieces that fill in as they go. What grows is the time: a 10-minute recording took about 40 seconds on a recent laptop's graphics card. On a phone, keep to shorter files: an hour of sound takes about 230 MB of memory.

Can I edit the transcript?

Yes. Click any line and type: fix a name, take out a filler word, rewrite a sentence. The text you copy and the TXT, SRT and VTT files you download all use your edits, and edited captions keep the timing of the line they're in.

Can I put the captions into the video itself?

Yes, with the free caption generator on this site: it transcribes the video the same way, then burns the captions into an MP4, word by word, in a look you choose. It can also burn in an SRT or VTT file you made here.

Is this video to text converter really free?

Yes. There's no price, no trial, no account and no limit on how many files you do. The video to text converter runs on your own device, in your browser, so it costs us nothing to run and we don't charge for it.

Is my video or audio uploaded anywhere?

No. The file is opened and processed by your browser, on your device, and never sent to a server: not to us, not to anyone. You can check: watch the Network tab of your browser's developer tools while it works, and you'll see the file is never sent anywhere.

More free tools for your videos

Silence remover

Cut the pauses and dead air out of any video or recording, automatically. Full quality, no watermark.

Add captions to video

Auto-captions burned into your video, word by word, in five looks, clear of TikTok, Reels and Shorts buttons. Nothing uploaded.

Filler word remover

Find every um and uh in a video or recording, check them in the transcript, and cut them out. Free, no watermark.

Video compressor

Shrink any video to a target size or quality as an MP4 that plays everywhere. No upload.

Video resizer

Change any video to 9:16, 1:1, 4:5 or 16:9 with a blurred fill or a crop. Full quality, no watermark.

Extract audio from video

Save any video's sound as MP3, WAV or M4A, the original copied untouched where possible.

Video trimmer

Cut the start and end off any video, frame-exact or lossless with no re-encoding.

Loudness normalizer

Bring any audio or video to −14, −16 or −23 LUFS, peaks held under −1 dBTP. Video untouched.

Apply LUT to video

Grade a video with your .cube LUT or a free look, see it live, and download the MP4. Nothing uploaded.

TikTok safe zone checker

See TikTok's, Reels' and Shorts' buttons and captions over your video, and download 1080×1920 PNG templates.

Auto reframe video

Landscape to vertical with AI face tracking: the crop follows the speaker, smoothly, shot by shot. Nothing uploaded.

Blur faces in video

Every face found and followed automatically, then blurred or pixelated. Keep the ones you choose. Nothing uploaded.

Audiogram generator

A clip of your episode as a video: cover art, a moving waveform and word-by-word captions, in four shapes.

MP3 to MP4 converter

Any MP3 as an MP4 with your image on it, the whole file, ready for YouTube. Fast, even for an hour-long episode.

Bleep out swear words

Find every swear word in a video or recording, check them in the transcript, and bleep them out. Free, no watermark.

Remove background noise

Take the fan, hum, traffic and wind out from under the voice in a video or recording. Hear it before and after. Free.

Teleprompter

Read your script to camera: scrolling at your pace or your voice's, mirrored for glass, and recorded as an MP4.