Kokukeyboard starts as a dictation app: press a hotkey, speak, and the text is typed at your cursor
in whatever program has focus. Pro adds a second tool for audio you have already recorded.
Instead of one take at a time, you hand it a whole folder, and it transcribes the files in turn while
you do something else - or while you sleep.
Each file's language is detected automatically across 90+ languages, so a mixed archive does not
need sorting first. Every document is written beside its recording rather than into a separate
library, so the transcript and the audio stay together. Because the work happens on your own CPU or
GPU, there is no server queue, no per-minute meter and no file-size limit set by someone else's
service. The only limits are your hardware and how many hours of audio you give it.
Once the Whisper model is downloaded, Pro needs no internet connection to transcribe. A laptop on a
train, an office with patchy Wi-Fi, a field site with no signal: the folder still turns into text.
Pro is $69 once - everything in the standard license, plus batch
transcription, with no subscription. The Pro batch tier is Windows only. Batch transcription is included in the 14-day free trial, no
card required, and every purchase has a 30-day refund. For a longer look at who uses it and why the
no-upload design matters, read offline AI transcription for whole
folders of audio. If you only need everyday voice typing, see
offline dictation for Windows.