Usually, when I tell people I do speech synthesis, they immediately presume text-to-speech (TTS), and make very fast judgements about The Festival Speech Synthesis System. Well, I do speech synthesis differently than most people. I use phonetics to generate synthesized speech. For over 25 years, phonetics has been part of my own self-studies in the field of linguistics.
That image is Festival in action, on my BeagleBone Black. The "saytime" command is a BASH alias for the flite command to announce the time. I'm not able to record video at this time as my camera seems to be putting lines in the pictures I take. Here are the phonetic conversions for the Tagalog language:
To work the demo, sudo apt-get install festival has to be done in Ubuntu.
Here are the voices I have already installed onto my BeagleBone Black:
Today is day #5 for having the BeagleBone Black. I've had a lot of trouble in getting it to work with my 22-inch Xenon TV, so I decided to settle with a headless mode setup. This allows me to work my BeagleBone Black using just an SSH connection. Thanks for reading and have a nice day!