Does anyone think that this would be possible to port to Pokitto?
You had me worried there. Usually, when someone says “Speech?” it means I have to stand up and make a fool of myself in a social occasion.
how good/bad is it?
A bit more robotic than the intellivoice, but plenty of documentation for making it sound different.
Intellivoice had 2kb of samples in ROM, not sure we can do that.
I did that with a midi converter about 20 years ago, fun, but terrible.
Problem is it’s abandonware The only open speech synth I know is espeak. It’s pretty compact, but still might be too complex for Pokitto (haven’t studied it). How about writing a super tiny original synth for Pokitto, and make it port-friendly for other platforms too?
There are techniques for sampleless speech synthesis right? (I don’t have much knowledge in this field, but from learning about vocoders at school I remember some speech models based on resonators that basically only require storing a few coefficients?)
Pretty sure someone ported this to Arduboy, so I’d say by logic of processor power that the answer is ‘yes’.
If not, just port this to Pokitto: