Does anyone know by what mechanisms the Software Automatic Mouth (SAM) that is used for text-to-speech on machines including Commodore computers synthesized speech? The manual for the product mentions something regarding target frequency spectra; does this mean all phonemes are generated just using additive synthesis?
In experiments I have done in the past, I have had some success synthesizing vowel sounds using formant synthesis, so I would imagine vowels could be similarly synthesized using additive synthesis, but then by what mechanism are consonant phonemes synthesized?
I have not had much success in finding answers to this from searching online, so I am hoping someone has an explanation of how this product synthesized phonemes.

  • The code is available at github.com/s-macke/SAM so you can see exactly how it works. – Mark Aug 23 '19 at 14:00
  • It looks whoever ported this or wrote the readme and comments on it didn't fully understand how it worked, either. Lots of variables have seemingly meaningless names, and the explanations given appear to just be speculation. – user2649681 Aug 23 '19 at 14:04
  • If you read the background on this it was a conversion from assembly to C – Mark Aug 23 '19 at 14:06
  • This sounds a better fit for dsp.stackexchange.com – Tetsujin Aug 23 '19 at 14:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.