There are only 3 Filters being used at one time. The 10 different settings are for individual vowel sounds.
Yeah I agree. Not exactly happy with the lookup table I used. Tried to adjust to taste with the resonance settings (which I incidentally hid from view in the patch I uploaded), but it still didn’t quite pan out.
The smoother (slew limiter) controls the attack/decay of the Formant slider controls. So if you record a parameter change at one step and you change at any subsequent step, the slew limiter smooths that transition out. Not sure why it would bounce for you. Perhaps turn up the rise/fall values? 🤷🏼♂️
Use that Formant table from Uni Munich
It sounds very good
I use it all the time
It’s average frequencies over 100 ppl or something if I remember right
Fixed some of the basics
Use hidden mixer at end to switch
Chose male to continue
Can’t attach stuff in private messages…
U a y work already so u see how it works …
Enjoy
Lala, you in Discord? There’s a Patchstorage discord server that would make collaboration/conversation on anything flow a little smoother: https://discord.gg/deByAcxk
here are the other formants
ü,ö,ü,y
m=male
w=female
ch=child
Attachment Bildschirmfoto-2020-11-26-um-14.21.37.png
oops
ü,ä,ö,y
not sure why you need 10 filters?
its just 4 frequencies
formant table is from linguistic studies university
Looking at the patch u only use 3 instead of 4 frequency
4 sound better
I guess you didn’t have the right lookup table
Not sure what the smother is supposed to do?
What I hear just go’s plong plong straight from State a to b …
btw. I would feed it with supersaw so you get a nice thick stereo sound from the start
oh and I would add a lowpass filter after the layer
formants have nothing above 4000 hz
I guess that’s enough inspiration for today :)
Used pulse width waves at the advice of: https://youtu.be/7tP3i6ST1z8
And the lookup tables were derived from: https://cnx.org/contents/nTOPtfsM@2.5:e8NdvtVS@2/Formant-Vowel-Synthesis
There are only 3 Filters being used at one time. The 10 different settings are for individual vowel sounds.
Yeah I agree. Not exactly happy with the lookup table I used. Tried to adjust to taste with the resonance settings (which I incidentally hid from view in the patch I uploaded), but it still didn’t quite pan out.
The smoother (slew limiter) controls the attack/decay of the Formant slider controls. So if you record a parameter change at one step and you change at any subsequent step, the slew limiter smooths that transition out. Not sure why it would bounce for you. Perhaps turn up the rise/fall values? 🤷🏼♂️
Thanks for the feedback.
For anyone else studying this stuff here a link to a study that cites the origin of the image lala posted (Bildschirmfoto-2020-11-26-um-14.21.37.png): https://www.phonetik.uni-muenchen.de/studium/skripten/SGL/SGLKap2.html?hcb=1
The original Formant table data came from: PETERSON, BARNEY, 1952, p.183
Use that Formant table from Uni Munich
It sounds very good
I use it all the time
It’s average frequencies over 100 ppl or something if I remember right
If you want to finetune it adjust level of frequencies
Fixed some of the basics
Use hidden mixer at end to switch
Chose male to continue
Can’t attach stuff in private messages…
U a y work already so u see how it works …
Enjoy
Attachment Formant-filter-fixed-basic-chosse-male-to-continue.zip
Thanks!
I like collaborative patching, it’s much more fun. :)
Lala, you in Discord? There’s a Patchstorage discord server that would make collaboration/conversation on anything flow a little smoother: https://discord.gg/deByAcxk
No, but we can do ping pong here.
If u fix the Formant switch that I messed up, I would continue
Or we can take it to drambo forum,
My current online home
I’m working on it slowly. Kids and all 😅. A couple formant frequencies got hosed too. I’ll check something in when I can 👍
😎