Read what was said.
On-device transcription gives recordings a written form to return to alongside the audio.
I’m developing an independent iPhone app for working with Limitless Pendant recordings. A place to listen back, read what was said and find the points worth carrying forward.
In development iPhone
The story behind it“The shelf fits, but the edges need smoothing before I paint it. I’ll pick up sandpaper tomorrow. There’s green paint in the workshop.”
A written example of a spoken note
The shelf fits and needs its edges smoothed before painting. Sandpaper is needed; green paint is already in the workshop.
The same thought, easier to revisit
Pick up sandpaper tomorrow.
I’m interested in what happens after something has been recorded. How do you return to the useful part, rather than simply collect more audio?
My work on Pendant covers the connection to the wearable, bringing recordings into the app and playing them back. The app also supports recording through the phone’s microphone.
Giving audio a written form makes it possible to return to what was said in another way. Transcription and summaries are processed locally on the phone.
On-device transcription gives recordings a written form to return to alongside the audio.
Multilingual transcription with automatic detection of the spoken language.
Summaries, key points and actions are processed locally on the phone.
From audio to a transcript, from a transcript to something useful.
A transcript is a starting point. A summary offers a shorter account; key points and actions help bring the next step into focus.
That connection is the part I’m working on: moving from capturing audio to understanding what was said and deciding what to do with it. Pendant is still in development.
I’d be happy to talk about the project and the ideas behind it.