Happy 9th birthday flower 
Profile posts Latest activity Postings Awarded medals About
-
oh :( lite version doesnt have vocal modes. i guess thats makes sense but a little sad by that
not an unpopular onion i just had this come to mind when i glanced through-- but its interesting i never really see ppl think about or even theorize on why an AI voice would sound so different from its previous non-AI voice, because it is kind of typical for no-AI voices to basically reflect like.. The key sound of that person with no influence on how they may sound normally- talking or singing in an outside context that isnt short samples? thats basically why. i feel like the most blatant comparison was Dex compared to his provider. theyre even featured in songs together and theres many parts where i can hear "dex" in Lupin, where dex as constantly reflects that unique tone through the song. at first they sound different until you pick it up in Lupin's voice.
with flower- i dont think Ci Flower was actually trying to be that different from v flower, have you heard her x voice samples? the spoken recordings? i remember getting those and thinking "wow that doesnt sound like flower at all" and they were the first example of what her VP sounds like outside of the vocaloid bank. Her voice samples sound WAY more akin to her Talk bank and later on Ci flower. this shouldnt exactly be a shock to hear because its been showcased before. Vocaloid flower took that unique tone that appears sometimes, and made it constant, because every individual sample made for a non-AI is way more likely to do so.
it happens with my own UTAU and im sure it happens with everyone else, as much as the bank still sounds like you. that is like, the exact reason any of these banks would sound more and less like each other- even a voice as unique as Teto managing to get the right sound, its still different and i feel the places where it is different are for that reason. the samples for AI voicebanks will never be as "exact" as the non_AI ones (they could potentially, with different or more data of some kind).
was listening to abuseken's stuff again as theyre probably one of my fave artists and forgot in 2020 they posted a song and mentioned they had turned 18. now its no surprise to find such young artists of course- theres a lot of them in the voca fandom. i guess what hit me most tho was hearing the subjects and themes of their songs and realizing how young they are.... like sure id be jamming to that at 18 thinking how relatable it is but im older and things have gotten better. so i cant help but really feel for the younger artists putting out their raw feelings while still dealing with such issues actively in their life. i got older and i found safety and a better life so i hope that happens for all those kids out there venting into their art.
pretty much every v flower artist i follow has still been using v flower and not ci flower-- minus a couple real popular once who maybe did a release demo. this isnt to say its proof ci flower is "bad", i think it just shows specific versions of voices are preferred to certain reasons. (and also i imagine some dont want to drop money for a new engine and learn it too). most of these v flower artists, using flower for that specific sound and it works! and i cant really imagine a lot of their styles translating as well to ci flower. i think too many ppl see a new, and quite different voice as the "upgrade" everyone will have to move to when thats not the case.
was reworking a Tangled cover i did that had some light "talking" to see if the rap function could help out. its interesting, it definitely makes the attempt at talking easier i guess? the way the parameter works. the intensity of the talking tho (theyre like, yelling/shouting) is still hard to reproduce. but i think for neutral speaking this will be useful
im working on something i think Jun might be good for instead of yuma, mostly bc of the tension response i have heard from jun and also yuma does lack vocal modes :( i wish his "solid" was a bit more like idk, solaria's solid. i dont always feel like his solid does too much in a certain range. his voice is a lil too soft in spots for this cover i want to do.... but idk if i wanna buy jun without really feeling i can use him a lot
for some reason some of my praat files are crashing synth v... but some are not. unsure if its praat problem or synth v :( hope i can fix this issue. its random.
ahhhh i know i would want JUN but im just not sure if i should buy him rn >_> i always jump the gun and get voices asap. which is fine. but my collection is getting large and i guess i wanna make sure im even using what i have ya know.
i plugged ritchy into a cover i had sitting around w Asterian, a very much musical song that was half talk-ish... i struggled with feeling like Asterian fit even tho his voice was like the original and my god. in manual mode, Ritchy sings it well. his way of still being a rap vocal works SO well...
i joined Vi-control the forum that linked my cover, for when i make some breakdown synth v videos.... i feel so out of place lol. its not that weird but its just a very different crowd of music ya know. its producers looking for a whole range of "virtual instruments" and are *not* of the fandom variety. and its not THAT weird for me technically- i at least live around, grew up around, western musicians and producers (my dad works in music) so i get It enough that i know i can probably communicate the vocal synth stuff in this forum in a way that will... make sense. even just glances around the vocal synth sub forum in there is funny. they talk very differently from us about it. lots of terminology here isnt passed through, lots of concepts we apply arent known and they apply their own existing ideas and understandings onto it in a way we dont. its a whole diferent perspective
im remaking that solaria cover i made of Easy On Me with the new vocal modes parameter stuff, making it as accurate to adele's vocals as possible. i am still getting comments from ~normal people~ aka likely normal western music producers who are probably see my cover on some of those professional music forums it got to. so odd. the cover hasnt exploded in views by any means (compared to my others. its at 6.7k) but interesting to see random people pop up. i wonder if i can go to that one forum and post about my improved cover. i also plan to make some information videos breaking down that cover and talking about tuning and mixing since some of these people have been curious about it.
oh i just remembered how we were told who the VPs were for solaria and asterian. i guess we may find out for saros too? will be curious to see.
going through vocaloid covers i was listening too back when i was really getting into the program and remembered this spanish cover artist who did a lot of real good covers and i always loved their stuff with flower, esp this chandelier cover in spanish??? still really good imo
one of their first ones with flower was this and again, really good. i suggest taking a look at their channel for spanish covers!
i think the another flaw in Synth v's mareketing is not showing enough of the voice's ability for cross language. right now, the english selection is incredibly small for their library and most people completely unfamiliar with synth v would look and 1) have no idea cross lang is/may not notice is as a feature (they dont even have demos for it on voice pages) and 2) have no idea if any of those chinese or japanese voices even sound good doing english. of course i get the intent for the original language but, theres so many non-english voices i suddenly hear in english and think "oh man, i kinda wish i had them because they actually fufill the voice i want/i had no idea they sounded this good!". voices like Cong Zheng or An Xiao's english are pretty unique compared to what we have, and actually good, and again while they cant be advirtized as "english" because they are trained differently, the professional western crowd they keep wanting to aim for in america wouldnt even notice/care if it means they actually do have more options for english outside the small selection theyll see. they really gotta boost all their voices in an english context to american musicians if they want people to care more, because a bigger selection is a good look and would make people realize they actually do have a variety of options.
teto AI is very tempting to me aside from her as a character but because i do find utau tedious to use and i have often drifted from it in favor of synths like vocaloid/synthv/etc etc because theyre easier to use for projects.... utau is something i love for its concept and community creativity but if i had a more modern option i would leave it in a heartbeat lol... and so that being said i never used teto much at all and having a more streamlined way to use her sounds great
flower arrived early! Picture of the box:
I won't say too much until I have it all installed and can make a cover. But im excited!
got an email saying ci flower will arrive on april 4th :o hype
did it every come up what the winning prize was for the ci flower contest? like there was money and then a secret prize. and they tweeted about contacting the winner. but i dont recall if they ever said *what* it was. like itd make sense if it was a copy of ci flower but i would expect them to say that in order to encourage participants lol...
i just got my physical of solaria and asterian!! :D -
Loading…
-
Loading…
-
Loading…
-
Loading…