Oh! Hey there! Sorry about inactivity. I'm a little nervous now, aren't I? Haha...

I guess you can count this as an update to another post I made on Scratch from a few weeks back.

You might as well have noticed about a previous blog post I did back in September, which I ranted about RFI's usage of VocalWare, taking the voices from Oddcast. I'd just like to let you know that for my Italian-language writing, I'm using a VocalWare voice named Raffaele, which has a Genoese Italian accent which I like to let it read my own writing with, typically for parodies of existing formats from the RFI sound announcement manual.

But great news: that voice I used is exactly the original one you've heard at stations like RFI has! At least for the sound, which gets extremely close to the original, most likely. Some of the people who's ever waited to take a train in Italy know a voice like this, not as a person but of how it sounds like, and how I can replicate it because of VocalWare... unless, with a different code name. We call that voice Roberto, which is so essential it definitely got stolen by VocalWare.

As you can see, it's a faceless voice that a company, in which we call Loquendo, created back in 1997, and RFI has always wanted to cut their budget by changing the way announcements are played. Back in those days, all they had to promote that change was to do what they needed: they hire an announcer or a voice actor, just to speak the delays, platform changes and even train departures, other than stupidly appearing on a phone call in the Libero program on RAI Two.

Just then, everything has changed! As voice lines from the sound announcements manual are read by voices like Roberto, and as the innovative Information & Communication software comes into play, all-new AI-powered automation has emerged for traffic info and its changes. Unfortunately, Oddcast has stolen the voice from Loquendo, though without permission, but thankfully it allows amateurs like me to now experiment with that voice, and for me I can train that voice with test announcements like what I've wrote.

So, the conclusion? Although I'm not impressed by how cheap it is, I surely did enjoy hearing what speech synthesis programs were like from about 26 years ago. You can see the full spoken version on the Scratch website here.