Real-Time Phonetics Editing (like PRAAT)
N
Noxelius
An effect for the editing of the users phonetics would be a great addition.
A really good reference for that is the programm "PRAAT", which is developed open source under GNU GPL.
It works its magic only by algorithmic logic and without any AI/ML, thanks to the brilliant work of dutch Phonetics Scientists at the Amsterdam University.
In regard to "changing the age of a speakers voice" it works way better than any other voicechanger I have ever used.
(This program has a big influence on the software products made by Antares, I would presume.)
The official website for reference:
(Despite its looks, the program is still actively maintained!)
This would be an awesome feature to have as (near-)Real-Time, either as an effect in Voicelab, or maybe even as built-in voicechanger.
c
coke
Thanks so much for the suggestion! 🙌
I shared your idea with our Audio Lab team. Here’s what they told me:
They think that changing phonetics the way PRAAT does (actually editing the sounds of speech) can only really be done offline, not in real time.
PRAAT is a classic and super powerful tool, but most of its magic happens on recordings after analysis rather than live.
For formant shifting — one of PRAAT’s big features — we already use our own VoTrans tech under our SDK.
A real-time FX that changes phonemes (like /eh/ to /ah/) would be very experimental. We’ve tried similar things before (like accent conversion) but the results weren’t very natural.
In any case, we really appreciate your feedback — it’s exactly the kind of input that helps us decide what to experiment with next.
Keep it coming!