The sad state of text-to-speech in Linux just got less sad

I’m a lazy guy.  I read a lot while I surf the net, and I have a screen reader do all the work for me.  Currently, I have TextAloud mp3 do that for me on this computer.  It runs in a virtual Windows XP installation that I have in “seamless mode” in VirtualBox on my LXDE/Ubuntu desktop.  And it does a good job.  It’s got its quirks, but I much enjoy listening to the synthetic female voice… as it… uh… synthetic female…

Okay, back on track.  I also use a Linux alternative on another computer.  My first computer is running Ubuntu 8.04 with an LXDE desktop.  It runs a KDE (KDE fucking sucks, btw) program called “KSayIt” to handle text to speech.  The voice synthesis coming out of THAT is… shall we say… inelegant.

But while I was on Ubuntu Forums recently for another issue, I found a thread that I had started a long time ago, back on June 2, 2007.  That’s pretty much smack dab at the start of my Linux.Ubuntu experience.  Back then, I was asking about text-to speech options.  Since then, I’ve learned that the applications available for Windows are always much better and much more varied than they are for Linux.  Kind of sad to admit that, but the sheer number of Windows developers and users outnumbers by a huge margin the number of Linux developers and users.

So I’ve always found the state of text-to-speech software in Linux to be somewhere between non-existent and horrendously shitty.

Even KSayIt is a clumsy, slow, craptacular and muddled-sounding memory hog that has customization options to nowhere and the look and feel of unfinished software.

And KSayIt was only in the Ubuntu repositories up until 8.04, which is why that other PC of mine is only at Ubuntu 8.04 – never to be upgraded.

But hold the phone!  In that old thread I started back in 2007, I noticed a recent post from February of this year.  The poster talked about a GUI for eSpeak.  And again… Linux programmers have to be pushed into making a GUI for anything like it was the deep end of the swimming pool and they were in kindergarten wearing their “I ♥ Linus” water wings.

So INTO THE FUCKING POOL YOU GO, Linux programmers.  Now I have a GUI for a Linux speech synthesizer, which means that I can use it like I’ve been using one in Windows since the last fucking century.

Now… there’s another issue.  And that is that eSpeak sounds even WORSE than the “Festival” voices I can use with KSayIt.  It sounds like a drunken Scottish C3PO.

Or worse, a Dalek.

I fear that if the word “exterminate” ever appears 3 times consecutively in something I’m reading, BAD THINGS will happen.

But at least I have a GUI option.  Which means I can probably go ahead and upgrade my very first PC to run the latest version of Ubuntu – of course, with LXDE replacing GNOME.  And when I get around to doing that, I think I’ll swap out the old and slow 10GB hard drive for a not so old and not so slow 40GB hard drive.  But that will be a project for next weekend, or whenever I’m not so exhausted that I need to play video games and drink tea all weekend.

I won another game of Civilization 2, by the way.  While listening to Mr. Drunken Scottish C3PO read about Russian history.  Nerd fun for a hardcore nerd.

Here’s another Venn diagram: