Re: NVDA speech synthesizer servay
Hi
I also chose number three, but I listened to each one several times at different volumes, and did them all apart, then close together as well, and in the end, it was kinda a choice between complete crap, crap, and not complete crap.
Yes, I know they put time into it, yes, I know it's multilingual and supports different dialects, and that each one is fast and accurate and free, I appreciate all of that, but when I was honest to my self it was still the case.
Still, that one I could tolerate for far longer than ESpeak, even with all of it's different voice variants that I've tried many times each.
Biggest problem with number three, which is the same with number one of course, is that the ends of words tend to hang when they are at the ends of sentinces most, but also sometimes during small breaks, which isn't conducive to higher speech rates, I mean, the hangs would decrease in length, but it still sounds kinda s tupid, and more so for books and articles.
The British dialect was pretty convincing though, which was neat, and it was alright with distortion at higher volumes, but still not great.
I like how they got rid of allot of the Kermit E. Frog from number one, in number three, and the fact that a bit is still their isn't so bad, considering how flat it would be with out it.
I think the most horrible was number two, along with the voice just sounding stupid and Sapi4ish, it also sounded like all the words were leaning against each other for support, all falling out in an endless stream of winyness, and we can't forget the chronic case of mouth open too wide sindrome.
_______________________________________________ Audiogames-reflector mailing list Audiogames-reflector@sabahattin-gucukoglu.com https://sabahattin-gucukoglu.com/cgi-bin/mailman/listinfo/audiogames-reflector
