Transcribe Video to Text
Turn the speech in a video or audio file into text, then download it as an SRT or VTT subtitle file.
- Free
- No watermark
- No upload, runs on your device
Drop your video or audio file here
or click to browse
Supports video and audio files up to 2048MB
How to transcribe a video to text
1 Drop your video or audio
Drag a video or an audio file such as MP3 or WAV onto the drop area. It is processed in your browser and never uploaded.

2 Let it transcribe, then fix words
Whisper turns the speech into timed segments on your device. Click any segment to hear that moment and correct the text, or split and merge segments.

3 Download SRT or VTT
Press Download SRT or Download VTT. The transcript lands in your downloads folder with a timecode on every line.

What the transcriber does, and its limits
Drop a video or an audio file and a Whisper speech model transcribes it inside your browser tab. The model downloads once and is cached, and your recording is never sent anywhere. The result appears as a list of timed segments next to the video. Click a segment to jump to that moment, type to fix a word, split a long segment or merge two short ones.
The model on this page is English only: base.en on desktop and the smaller tiny.en on phones. Clean speech from one speaker comes out best. Background noise, people talking over each other and strong accents lower the accuracy, and there is no speaker labelling, so a conversation reads as one transcript. On a laptop, expect transcription to take roughly one to three times the length of the recording. Files can be up to 2 GB.
There is no plain .txt download. For plain text, select the transcript and copy it, or open the SRT in any text editor. For other languages, the text-based video editor Orator has a multilingual model on desktop, and it lets you cut the video by deleting words from the transcript.
SRT and VTT subtitle files
SRT is the format almost everything reads: YouTube Studio, Vimeo, Facebook, Premiere, Final Cut and most video players. Each caption block is a sequence number, a time range written as HH:MM:SS,mmm, the caption text, and a blank line. The file is plain UTF-8 text, so you can open and fix it in any editor.
VTT, short for WebVTT, is the format for HTML5 video on a website you control. It supports positioning and styling cues that SRT leaves out. One transcription run gives you both files, so there is no need to choose before you start.
Whisper places segment times to about a tenth of a second, and a boundary can sit a few tenths away from the spoken word. That is fine for captions. If you need frame-exact sync, adjust the times in a subtitle editor afterwards.
When this fits
Show notes and quotes
Pull an exact quote from an interview or write show notes from a talking-head video without listening through it again.
YouTube captions
Upload the SRT in YouTube Studio next to the video and the CC button works for every viewer, on every device.
Editing in Premiere or Final Cut
Both editors import an SRT as a subtitle track. Drag the file onto the timeline and the captions land as editable text.
Translation
A translator works through an SRT one block at a time. Generating the English SRT here gives them the timecodes already in place.
Frequently asked questions
Can I transcribe a video to text for free?
Yes. There is no signup, no per-minute charge and no limit on how many files you transcribe. The only limit is what your device can hold in memory, up to 2 GB per file.
Is my file uploaded to transcribe it?
No. Transcription runs in your browser on your device. The only thing fetched is the speech model, which is cached after the first download. Your recording never leaves the tab.
Can I download the transcript?
Yes, as an SRT or a VTT file, both with timecodes. There is no separate plain text export; select the transcript and copy it, or open the SRT in a text editor and remove the numbers and times.
Can I edit the transcript before downloading?
Yes. Each segment can be edited, split at the cursor, merged into the one before it, or deleted. The downloaded file reflects whatever is in the editor when you press Download.
How accurate is the transcript?
Clear English from a single speaker usually comes out with few mistakes. Background noise, overlapping voices, strong accents and specialist terms lower the accuracy, which is why the editor sits right next to the video.
What languages does it support?
English on this page. For other languages, Orator, the text-based video editor, has a multilingual model on desktop that detects the language on its own.
Can I transcribe audio only, without a video?
Yes. Drop an MP3, WAV, M4A or other audio file and it is transcribed the same way.
What is the difference between SRT and VTT?
SRT is older and read by almost every platform and editor. VTT adds styling and positioning cues and is the format HTML5 video players expect. For YouTube, social platforms and editing software pick SRT; for your own website pick VTT.
Related tools
Auto caption generator
Style the captions and burn them into the video.
Text-based video editor
Cut the video by deleting words from the transcript, with more languages on desktop.
Extract audio from video
Save the sound as MP3 or WAV, for example to send to someone else.
Remove filler words
Clear um, uh and like from a recording using its transcript.
Burn a subtitle file into video
Put a finished or translated SRT onto the video.