Festival 1.96 text to speech. Festival 2.x available in modern distros does not support the 1.9 voices and the Festival 2.0 voices are far inferior. There are no comparable replacements for native text to speech on linux. Just a bunch of cloud stuff and bleeding edge things that require nvidia GPUs, etc. If an OS can't run Festival 1.96 then it can't be my main desktop.
imgSeek. It requires python 2.4 among other things. It allowed me to draw crude MS Paint style images and then use them to search for similar colored/shaped images on my hard drive. There's no good replacement for it's ability and porting is non-trivial. https://sourceforge.net/projects/imgseek/
Also, the relevant xkcd: https://xkcd.com/2224/ My solution for this is to keep my old computers running old OSes and software. I build a new computer with all the new software once every 5-10 years when I'm finding I can no longer compile things.
I assume you mean the CMU ones? They're decent for quick stuff but listening to a whole book with them is unpleasant. They don't really compare to the Nitech voices.
Comments
Festival 1.96 text to speech. Festival 2.x available in modern distros does not support the 1.9 voices and the Festival 2.0 voices are far inferior. There are no comparable replacements for native text to speech on linux. Just a bunch of cloud stuff and bleeding edge things that require nvidia GPUs, etc. If an OS can't run Festival 1.96 then it can't be my main desktop.
imgSeek. It requires python 2.4 among other things. It allowed me to draw crude MS Paint style images and then use them to search for similar colored/shaped images on my hard drive. There's no good replacement for it's ability and porting is non-trivial. https://sourceforge.net/projects/imgseek/
Also, the relevant xkcd: https://xkcd.com/2224/ My solution for this is to keep my old computers running old OSes and software. I build a new computer with all the new software once every 5-10 years when I'm finding I can no longer compile things.
Check Flite2 with the voice pack. A lot of them like SLT and RMS (not that RMS) are good enough.
I assume you mean the CMU ones? They're decent for quick stuff but listening to a whole book with them is unpleasant. They don't really compare to the Nitech voices.