arisuchan    [ tech / cult / art ]   [ λ / Δ ]   [ psy ]   [ ru ]   [ random ]   [ meta ]   [ all ]    info / stickers     temporarily disabledtemporarily disabled

/tech/ - technology


formatting options

Password (For file deletion.)

Help me fix this shit.

Kalyx ######

File: 1499283694011.jpg (237.38 KB, 1920x1080, tmp_9174-Vocaloid-Wallpape….jpg)


How far has speech synthesis come along in recent years? Is there any software that sounds better than the ivona-esque monotone stuff that we've all seen before? Vocaloid has gotten rather impressive over the years, but is there anything that can convincingly mimic the intonation of casual human speech? I'm interested in playing around with it a bit, but from what I've read before the development seemed a bit out of my league, so I was rather discouraged.


as you've mentioned, the tech has been here for lots of years already; the problem is that handling all the little case-by-case changes and idiosyncrasies requires giant training sets, only giant companies can afford to make those, and they keep everything in-house.

so no, there's not really anything publicly available; would require some kind of impossibly huge crowd-sourced training effort

[Return] [Go to top] [ Catalog ] [Post a Reply]
Delete Post [ ]