Ploki

Wish to replicate it on max. Not really sure how IBM 704 did it, so don't know how to approach it.

https://www.youtube.com/watch?v=ebK4wX76RZ4

https://www.youtube.com/watch?v=5hyI_dM5cGo

Wish to replicate it on max. Not really sure how IBM 704 did it, so don't know how to approach it.
not sure about the Voder either.

https://www.youtube.com/watch?v=ebK4wX76RZ4
https://www.youtube.com/watch?v=5hyI_dM5cGo

daisy-style-ibm704-speech-synthesis-voder-style-speech-synthesis

Assuming you know about formants, and voiced & unvoiced sounds, you will learn a lot by studying the standard Max example 

classic_vocoder effects -> classic-vocoder-folder.

Assuming you know about formants, and voiced & unvoiced sounds, you will learn a lot by studying the standard Max example classic_vocoder effects -> classic-vocoder-folder.


I thought there was more to daisy than just vocoding :( I thought there was some real formant-synthesis involved.

https://ccrma.stanford.edu/~jos/pasp/Singing_Kelly_Lochbaum_Vocal_Tract.html

if you're into that kinda thing, the sonivox is cool too, it's similar to talk box. 

https://www.youtube.com/watch?v=blhqeWpwwkA

seen this? https://ccrma.stanford.edu/~jos/pasp/Singing_Kelly_Lochbaum_Vocal_Tract.html

if you're into that kinda thing, the sonivox is cool too, it's similar to talk box. https://www.youtube.com/watch?v=blhqeWpwwkA


For Daisy, I think they used a source-filter synthesis, with 2 formants defining the vowels.

I gave you the Moog-like vocoder reference because the synthesis algorithm is basically the same.

Difference is that in the Moog imitation, the "formants" are fixed: actually a bank of 15 filters or so.

For Daisy, I think they did use 2 moving formants to get the vowels.

Ploki,
For Daisy, I think they used a source-filter synthesis, with 2 formants defining the vowels.
I gave you the Moog-like vocoder reference because the synthesis algorithm is basically the same.
Difference is that in the Moog imitation, the "formants" are fixed: actually a bank of 15 filters or so.
For Daisy, I think they did use 2 moving formants to get the vowels.
Jean-François.


Daisy style (IBM704) speech synthesis, "Voder" style speech synthesis