Video to text
Drop in an MP4 or a MOV — a lecture, a talk, an interview, a screen recording — and get the words out with timecodes. No audio to extract first, no upload, no per-minute meter. Subtitles export as SRT or VTT on the way out.
macOS 14.4 or later · Apple silicon · $29 once, not per month · 7-day trial
How it works
The video stays where it is. What moves is the text, once there is some.
Drag an MP4 or MOV onto the Transcribe screen, or right-click it in Finder and choose Open With → Kekoso. There is no step where you extract the audio first — the video file itself is what you hand over.
The speech model runs on your own machine, on the audio inside the container. A long recording keeps going while you do something else; nothing is queued on a server and nothing is metered.
Copy the text, or export SRT and VTT with timecodes and load them straight into your editor, your player or YouTube. TXT is there when you want the words without the timings.
What the video transcriber does
Most ways to convert video to text are a website with a file size limit and a free tier that ends on the second upload. This video transcriber is an app on your Mac, and the limit is your disk.
The two containers almost everything produces — phones, screen recorders, editors, cameras. Kekoso opens the file and works on the audio track inside it, so there is no ffmpeg step and no intermediate WAV to clean up afterwards.
Every transcript carries timecodes, and SRT or VTT export turns a recording into a subtitle file for the video it came from. Cues are grouped by broadcast readability rules rather than cut wherever the model paused — the details are on the subtitle generator page.
A three-hour conference recording transcribes the same way a ninety-second clip does. No plan tier, no per-minute billing, no upload limit — the video never goes anywhere, so there is nothing to meter.
Point Kekoso at the folder your screen recorder or camera saves into, and each new video is queued as it lands. A day of footage becomes a day of transcripts without a single drag.
Paste a link to a video you own or have permission to use, and Kekoso fetches the audio and transcribes it the same way. That fetch is the one network request in the whole feature.
Parakeet for fast English and 24 other European languages, Whisper for the long tail, SenseVoice for Chinese, Cantonese, Japanese and Korean. Each model card shows measured accuracy and speed, so the trade-off is yours to pick.
Where the video goes
An online video transcriber starts by asking for the file. That means a copy of an unreleased cut, an internal all-hands, a client walkthrough or a recorded consultation sitting on a server you do not control, under terms you probably did not read to the end.
Kekoso reads the file in place. The transcription pipeline makes no network calls, nothing is copied, and the transcript is a plain local file. Turn off Wi-Fi and transcribe a video: it behaves exactly the same, because there was never a server in the loop.
How to verify that yourselfWhat it does not do
MKV has no registered file type on macOS at all, so nothing can identify it; WebM and WMV announce themselves as playable video and then refuse to decode. This video transcriber tells you which of the three happened instead of failing quietly. Converting the container first — one pass through HandBrake or ffmpeg — makes the file readable, and that conversion is not something Kekoso does for you.
A film bought from a store, a lecture inside a course platform, anything with protected content: the audio is not readable by any app that is not the store’s own player, and Kekoso says so rather than producing an empty transcript.
A screen recording made without capturing audio has no speech in it, and no software recovers what was never recorded. Kekoso detects the missing track before queueing the file, so you find out in a second rather than after a long wait.
A conference talk or an interview filmed on one camera comes back as one continuous transcript with timecodes, not split by who was talking. Calls recorded on the Mac itself are the exception: your microphone and the other side are captured separately and labelled as such.
macOS 14.4 or later · Apple silicon. Recognition leans on the Neural Engine, which Intel Macs do not have.
Questions
Pricing
Four jobs that usually arrive as four subscriptions, running on hardware you already paid for.
7-day free trial. No card to start.
macOS 14.4 or later · Apple silicon · payment handled by Lemon Squeezy
7 days, every feature unlocked, no card and no account. If it does not earn its place in your week, delete it and nothing follows you.
macOS 14.4 or later · Apple silicon