• We're currently having issues with our e-mail system. Anything requiring e-mail validation (2FA, forgotten passwords, etc.) requires to be changed manually at the moment. Please reach out via the Contact Us form if you require any assistance.
morrysillusion

morrysillusion

v flower enthusiast 25 From Socal
Messages
777
Reaction score
4,386
Medals
5
Joined
Last seen

Profile posts Latest activity Postings Awarded medals About

  • morrysillusion
    gotta say stardust's english really has a great sound. its not something id use but i feel like it falls in line with the young feminine pop vocal that a voice like Anri has, but different enough to still be unique from her. She does seem to have some issues with some consonants but idk i guess you cant complain when shes not supposed to speak english technically.
    • Like
    Reactions: WyndReed, Blue Of Mind and MillyAqualine
    morrysillusion
    morrysillusion
    i do feel like japanese/chinese voices singing english have.. and interesting advantage? i think they often sound a lot more "relaxed" with their pronunciation liklely due to the fact they have limited sounds to work with compared to a native english bank. while Stardust shows some issues due to it, i think i have really enjoyed how they sing bc they dont have the same incredibly strict pronciation as most english synths. i think synth v's english banks have improved off a lotta things, but i still feel voices like eleanor, anri and kevin just like... have that borderline "sing like i need to enunciate every word super strongly" problem that is more apparent in english vocaloids. they sound great but, i would rather have relaxed singing, even if it comes out a bit unclear sometimes, and it interesting i feel the nonenglish voices acheive this at times bc of their less sounds.
    Prism
    Prism
    What I do find really interesting about stardust's cross lingual english is that she sounds very close to her vp when she sings in english. I never really heard it in her vocaloid or when she sings chinese so I find it really interesting. Native english synth Vs may I suggest using voicing it helps a lot with relaxing the voice. I use it a lot in pop songs were there's slurred pronunciation. Also can help for when you want to easily change the emphasize of a phrase.
    morrysillusion
    morrysillusion
    tbh idk if its just certain voices but, Voicing has always just sounded more like its changing the volume to me and making them breathy, but i think its bc i know the tone im talking about is really just inherent to the actual recordings and not something the engine can change unless like, AI learning also includes it-- it can help in some ways but for me its not enough to be the change id want in the voice, as nothing will truly beat actually having a voice that is actually singing relaxed versus one being manipulated by a parameter. i also just really dont like the "breathy" grainy sound parameters like that put on the voice :(
    morrysillusion
    • Like
    Reactions: poser, pico, WyndReed and 3 others
    morrysillusion
    morrysillusion
    This is jokes bc there may be more to learn about the subscription model but regardless. Subscription model is not something I want to deal with in any form w vocal synth I think. And if it's "temporary" until you pay it off then... it's not a subscription lol.
    morrysillusion
    morrysillusion
    Truth tho, if gynoid got itself together for a cevio pro release I think technically, it would benefit them more than a full price release. Whatever is going on over there doesn't seem good and I'm not sure a full $100 price tag will help them in the long run as we all know how poorly voice synths actually sell. So.... I guess it wouldn't be a bad model for them in the end.
    morrysillusion
    tbh i am starting to think i am tone deaf or something. i try time and time again to make midis from reference, or fix a messed up melody, and i never ever get the notes right. i might get one, but its just constant fiddling. maybe i just cant get it right because the original voice doesnt sound like the synth or piano sound "enough" for me to think its actually correct? idk. its a shame bc im around so many musically inclined ppl, i love music, i know what sounds good! but referencing audio to make midis by ear? just doesnt happen. the only reason i have successfully made original music isnt bc i did it by ear- i only had stuff sounding right because i was constantly referencing a list of notes of a scale i chose, so.
    Mika
    Mika
    I feel you so much. I can tell when something sounds right or wrong and I'd say I'm also not that bad of a singer so definitely not tone deaf. But making midis??? Seems completely impossible to me, idk how people do it it's such a mystery to me
    Prism
    Prism
    morrysillusion
    i cannot seem to export my existing tuned files into a format that keeps pitch bends in cevio pro... so. thats a shame. i did a test render of two songs i covered with flower a while back, just as a fun comparison w her tone and sound in that style. so sadly theres no tuning bc of the conversion so :/ but take it anyways, flower on harmonies.

    Fuck you PTA - this is a very basic file, little issues here even without tuning, but she gets tripped up on the short notes in the chorus.
    Otome Dissection - this one was odd. it did not want to deal with note bending and sounds quite choppy due to it. i would cover something else but its late and im tired, might as well use the file as an example if you convert files like that for cevio lol

    heres links to the flower versions of these covers i did... obvs i dont think flower's vocaloid voice sounds like chis-a, it was more so her talk voice. but i think their voices will go very well together tbh and its interesting to compare when people obviously thought Chis-A could be flower's talk VA.
    morrysillusion
    anyone have artists theyd recommend for commissions of full body humans- notably for a utau design? whether its you or someone else- i just dont follow a lot of artists in the community and know i want to commission some art for my utau once i have the design down!!
    morrysillusion
    morrysillusion
    one of the issues about being a furry for ages and not indulging in the vocal synth community directly is i follow next to no artists who draw humans... so please if yall have suggestions on who to follow let me know lol. people who regularly do commissions, no matter how expensive- link them!
    morrysillusion
    i havent really seen anyone mention the um.... incredibly bad sounds synth v makes on lower pitches and man, it is terribly obvious now with voices like kevin and ryo. when i use all the fem voices they never need to be lowered really, but with ryo i... expect he can handle being a bit lower ya know. and i thought this was at first a glitch with him- but it applies to all the voice- holy crap, synth v. even if vocaloids dont sound "good" lower, they act least dont do whatever.... synth v does. to be honest its probably the biggest disappointment after having gotten ryo. i have to be very careful with his low range because even just one not gets all weirdly fucked up when its one note too low :/ i could kinda hear it in some of those Muxin demos too but thought it was just him.

    what i mean, is.... this sound. low voice weirdness the low range completely warps, sometimes cutting off and becoming weird and fuzzy. at first it was subtle and i didnt mind his 'reaction' to his whispery low range, until it sounded so strange i realized it was in fact not normal lol. and again, this isnt just him (anri, ele and solaria are all the same), its just how synth v...works i guess.
    • Like
    Reactions: Blue Of Mind, MemelordEnjoyer, WyndReed and 2 others
    morrysillusion
    morrysillusion
    like i said this really an annoyance because even vocaloid comes out synthesized in a way thats passable- this just... isnt. im surprised i havent seen more complained but i guess with the huge range of female voices, you wouldnt run into the issue because they just dont work low anyways. idk if this happen with non-AI voices like Genbu, but there are obviously a huge lack of male voices to where this probably didnt come up for most users anyways. and now with two new male voices? yeah, im a little unhappy to hear that tbh
    morrysillusion
    the test i did using Praat for pitch and timing is actually...really neat. i used to us vocalistener sometimes in vocaloid but that plugin is quite dated and doesnt have a great interface... praat definitely has a *lot* of stuff to offer but for what i learned for doing what i did it was a lot more intuitive, better work flow. so i may be trying this on some other songs i wasnt satisfied with due to not capturing the original singers vibe. idk if id do it constantly bc i feel at that why point "why" is the question- as its just copying the singers. but its neat to hear, and also as a tuner, to see more directly how the singer sings stuff
    morrysillusion
    i had some annoyances when i would convert a vsqx to svp with utaformatic and synth v would turn all default "Ooh" notes with [ - ] to "uw" because it wasnt ready the - phonetic in vocaloid's notes. and i wished it would just apply the [ - ] to carry and id have to alter it manually. but then i realized i could edit the .svp (or vsqx) in notepad++ and replace all "Ooh"s in the file to -.... thus making Synth V carry all the notes like it should. i always forget you can read those files in notepad
    morrysillusion
    i keep posting my art at like the worst hours here :') but i finish stuff late and then i just... post it late, or else i forget
    morrysillusion
    i think im going to buy ryo, but im trying to hold off bc im spending on a lot of otyher more important things right now.... but soon. its been a while since we had more avatar-less voices and i kinda forgot how fun it was to see how people interpret them tbh, i realize i really like voices with no avatar bc of that too.
    morrysillusion
    i purely exist when any news related to flower pops up
    • Like
    Reactions: TheStarPalace
    morrysillusion
    morrysillusion
    still left with a lot of mystery to the Huge pause in gynoid/flower news, if that does end up being flower.

    either its flower and something weird delayed it, or they have completely reinvented flower's character. maybe new ownership over her? idk. regardless the development of "flower cevio AI" has been strange no doubt, where ever its going
    morrysillusion
    im probably not gonna be on the forum for a while or not super active- glad to post my covers but i just need a break again, but i may pop in to post some covers i do with solaria if i do more
    morrysillusion
    new covers up. learned what tone shift did. best parameter tbh.
    morrysillusion
    morrysillusion
    anyways my general Onion on solaria now is: i really like her, really she sounds very realistic and i love seeing that. i LOVE seeing AI voices come out. i feel like we will only have more diverse options that more they develop. its gotta be remembered this tech, mostly in the west, is very new. we're only just seeing AI english voices come out and i cant imagine what will come next.

    i think solaria fits a certain kind of style, like any existing (tho more often AI) voice but she has potential to be pushed to other things. the way her voice handles more extreme parameters is really good so i look forward to pushing that. shes a nice mature voice with good strength, but also a good mellow low tone too. she wont fit everything but no voice ever will, really. i prefer the way she handles her words over Eleanor and Anri, but sometimes Ele's tone comes in a close second when picking my final voice.
  • Loading…
  • Loading…
  • Loading…
  • Loading…
Top Bottom