Nov 2, 2022
inspired by this post: https://www.tumblr.com/karhs/668219948353683456/oh-what-so-he-can-get-his-ass-kicked-when-he
seen from China
seen from United States
seen from Lithuania
seen from Brazil
seen from China
seen from United States
seen from Malaysia
seen from Malaysia
seen from United States
seen from China
seen from United States
seen from Puerto Rico
seen from United States
seen from United States
seen from United States
seen from United States
seen from United States
seen from Türkiye
seen from United States
seen from China
inspired by this post: https://www.tumblr.com/karhs/668219948353683456/oh-what-so-he-can-get-his-ass-kicked-when-he
tfw you are a Telvanni Wizard's apprentice, but Master Neloth keeps all the fun Arcane Knowledge locked up, refusing to teach you hardly any of it... and your lockpicking skills are woefully underdeveloped :'(
This is a prototype for one of the bridge ambient dialogues. Lines are not necessarily final; but they are representative of the first steps in a new level of vocal performance I can get. I figured I'd share some audio because I'm honestly kind of in shock right now. I started to play around with Speech-to-Speech conversion. Essentially, the synth transposes my vocal performance, including pitch, duration and inflection, onto another voice. This allows me to "puppet" another voice. In this case, I performed Joker. The process is kind of janky and often, the resulting output requires tweaking - however, often much less tweaking than generating a take without vocal input. Doing it this way provides an extremely helpful "sketch" of a delivery that I can then manipulate. It works best when I do it to FemShep's voice, because hers is very similar to my own tone and natural accent. It helps if I try to do an impression of her when reading the lines, as the synth understands what to do with the data a little better. Although my duration data is useful for performance, I speak much slower than Green. (Did you know that, generally speaking, males speak faster than females?) so, I run most of his lines through Audacity to speed up the tempo without disturbing the pitch. The results are some of the most emotive lines I've managed thus far. Yes, it's still got the trademark synth fuzziness to it, but it sounds much less like text to speech now, and that's because it's not strictly text to speech!
Sure, mostly I'm making him sound snarky and cranky here. But crucially, he doesn't sound flat and that's incredible. I'll be revisiting some of the weakest line deliveries and implementing reads using S2S if I can make them sound better.
@villainpunk said this on a post and i thought it sounded like something a bethesda npc would say so here’s this, thanks to xvasynth
skyrim mod that makes the guards say this instead of ‘let me guess, someone stole your sweetroll?’
This post inspired by the tags left by @sneakygreenbean on my previous post: https://www.tumblr.com/noneedtofearorhope/699833737250258944/inspired-by-this-post
Tfw Master Neloth yells at you to wake up, complains "i HaD to wAlk iN 40 ft of asH to gEt to WiZard SchOol", then suggests you develop a caffeine addiction and proceeds to shill for his favorite product, Canis.Root.Tea™
Uploaded this + the previous Neloth Alarm here, in case anyone wants these silly clips lol, in mp3 and m4r (iphone) formats.
