LesskeysTranscribe a video locally on a Mac or a PC
To transcribe a video locally, add the file to Lesskeys: it takes the sound out of the video on your computer and turns it into text, with speaker labels. The video is never uploaded, and it works with the internet off.
macOS 13+ on Apple Silicon · No subscriptionWindows 10 and 11 · No subscription
Last reviewed
Add the video
Transcribe a video locally, step by step
- Open Lesskeys, choose Transcribe and add the video. MP4, MOV and M4V work; convert other formats first.MP4, MOV, M4V, MKV, AVI, WEBM and WMV are accepted; convert a file that will not open.
- Leave the language on auto-detect, or pin up to four.
- You set how many people spoke (2 by default).Say how many people spoke: pick 1 to 4, or enter a custom number.
- Start it. In Library, press play on a line, or its time, and the video starts there.
The first run downloads a small set of speaker files.
What it misses
What it reads, and what it misses
Lesskeys transcribes what people say.
- Text on the screen
- Describing the picture
- Editing the video
Music and crosstalk cost words, and a wrong speaker count gives wrong labels. A long video can end early: when Whisper stops early.
Record the screen
Record your screen, then transcribe it
Lesskeys can make the video too.
Open Record Call, choose the screen or windowthe display and press Capture with Video. It records your computer's sound and your microphone, so a video call is captured with both sides, and no bot joins it. Stop, choose Transcribe, and the transcript appears in the Library next to the video. Call recorder for MacCall recorder for Windows has the steps.
Tell the people you record; some places require everyone's consent.
Your computer
The video stays on your computer
A website needs the whole video uploaded first. Here the file stays on your disk.
- Nothinguploaded
- No accountneeded
- Offlineworks with the internet off
Test it with Wi-Fi off: transcribe a video without uploading it.
On a Mac
How long a video takes on a Mac
Only the sound is read, so the picture size does not matter.
In our test on one Apple M1 Max, an hour of speech took about five minutes with the more accurate model. Taking the sound out of the video and finding the speakers add time, and an older Mac is slower.
Source: Which Whisper model to run on a Mac. To record the screen, macOS asks for Screen Recording permission the first time.
On Windows
Video on Windows
Only the sound is read, so the picture size does not matter.
Windows speed depends on the PC, and we have not measured how long an hour of video takes there, so there is no figure. Taking the sound out of the video and finding the speakers add time. To record the screen, Windows asks for nothing beyond your microphone, and the recording is saved as an MP4 next to its transcript.
Questions
Questions about transcribing video
Can you transcribe video locally?
Yes. Lesskeys takes the sound out of a video file on your computer and transcribes it there. Nothing is uploaded, and it works with the internet off once the models are installed.
Does transcribing video require an internet connection?
No. Once the models are on your computer, adding a video and transcribing it needs no connection, so it works in airplane mode. The speaker models are a small one-time download, the first time you transcribe.
Can it label who is speaking in a video?
Yes. Lesskeys labels the speakers on your computer. You can set how many people spoke, and you can rename or merge the labels afterwards in the Library.
Can I jump to a moment in the video from the transcript?
Yes. The Library plays the video next to the transcript. Press play on a line, or its time, and the video starts there, while the video file is still where it was.
More