• 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.
Ronny777
Reaction score
306

Joined
Last seen

Profile posts Latest activity Postings Media Albums About

  • MIKU V6 AI When YAMAHA? :|

    Maribelle
    Maribelle
    This is what Yamaha intended when they made v6
    razelberii
    razelberii
    This actually sounds pretty good, amazing
    Ronny777
    Ronny777
    @Maribelle honestly, this tech is so janky. I have no idea what Yamaha was thinking not adding a way to modify the Vocalo file. By playing around with the program i found that its really good with breaths and humming. I just with there was a way to smooth out parts inside the vocaloid program. best i'v been able to do is lower the sound where is glitches.

    @razelberii I know right! If the tech can get some sort of refinement upgrade it would be a lot more fun to play around with. Right now the best I can really make are Shit posts that make me laugh.
    I know this might sound straight up stupid or maybe its just way too early in the morning and need to go to bed but SARAH sounds so much like Lady Gaga. Her tone sound so similar? It's so uncanny when i listen to the original song in one ear and my WIP in the other. i didn't use Vocalo changer this is just SARAH in V6 (with Vocalo Changer she sounds like a drunk) anyway don't mind me I'm just really happy to finally be able to work with a Vocaloid that can sing gaga songs!


    Bonus Vocalo Changer version
    I need to get this off my chest and know what others think about this. So I've made a small list of what I think of Solaria and SynthV:

    I think Solaria sounds too human. To the point that I can't tell if its just the program that sounds this way with tuning or if its how she sounds "plug and played". I haven't used her and I really don't know much about how SynthV and how the AI works.

    I've thought about it for a while and I think my problem with the technology is that it feels uninspiring.

    I'm speechless by the realism of Solaria to the point that I can't really imagine how to improve her the way I do with MIKU and the other Vocaloids I work with.

    I can't see her as a character. She just sounds like any real life singer. Even my mom remarked that her demos was perfect. She had to ask me to confirm that I was showing her a synthesized voice.

    It's also hard to come to terms with the fact that Solaria sounds the way I've been trying to get MIKU to sing and I don't think I'll ever be able to achieve that with Vocaloid.

    It's like the hearing equivalent of the evolution of video game graphics. I just hope Vocaloid doesn't get left behind on the english language side.

    I've put so much time in VOCALOID T_T
    • Like
    Reactions: WyndReed
    morrysillusion
    morrysillusion
    yeah i get you-- i kiiinda mean both lol. character as in image, but also like. 'character voice' as in voices acted to be played as a character (ex rin/len) thus the uniqueness of the voices. tho i didnt specify as clearly
    lIlI
    lIlI
    If you'd have told me as recently as last year that we'd reach a point where vocal synths were criticised for being too realistic, and too authentic a capture of their voice provider, I wouldn't have believed it! That's pretty cool, in a way. As for unrealistic voices, I wouldn't worry about us lacking them any time soon: we have 20 years worth of unrealistic vocal synths and only one or two realistic ones. I'm pleased we at last have a couple of easy to use voices, since it has the potential to bring in a lot more original producers to the fandom. (For someone focused on songwriting and production, a vocalist that takes hours to edit is a very offputting.) I think for future voices that retain the classic vocal synth sound, UTAUs will continue to have passion poured into new releases.

    In terms of Solaria's editability, I surprisingly found her the most flexible vocal synth I've ever tried when it comes to turning her into different characters. She can become a very convincing masculine voice, handles both cute and mature performances well with a light touch of gender factor, and can go as soft and whispery as Eleanor with tension. I have high hopes for the future of flexible tuning, as it seems AI has increased how far you can push a voice while retaining quality (since it's not constrained by sample data). A good example is comparing Solaria in Circus' Lower cover vs Buy Buy Buy.
    Overall, I'm really happy that a frontier has been broken and made more options available! :chika_ani_lili:
    I
    inactive
    Speaking (well, typing...) as a non-tuner, Solaria is fantastic. But I still hope that something better comes along!
    Chris is so hard to tune! :rin_lili::rin_lili::rin_lili: His sounds are so inconstant. He sounds whiney in one place then sounds phlegmy in another then he decides. He wants to add perfect growsl to some notes that I can't replicate anywhere else that I need them. I swear he feels sentient. AI without the AI tech.
    I just look through my WIPs folder...... I've accumulated 418 V5 cover projects (20.66GB) from Apr 11, 2018 - Jan 14, 2022. (I'd say 20% percent of that is multiples of the same songs but using different Vocaloids and 80% MIKU.)

    I'm starting to wonder just how much more practice I'm going to need until I'm able make something that's worth sharing. Even when I feel like sharing I keep getting the feeling that I'll be spamming stuff that no one wants to listen to. It doesn't help that I'm scared of not being self aware enough to tell if I'm bad at tuning or not and making a fool of myself about being proud of something that sounds awful.
    MacneNana's voice is so perfect for covering Tones and I:petit_lili:. Too bad my tuning still needs to improve before I can do her any justice. So I'll share just share the only part Of my WIP I'm proud of:LOL:
    Every day or so since march 25 at 10AM MANDARAKE Likes to send me an email to tell me that the FuRyu BiCute Bunnies Figure WHITE Ver. HATSUNE MIKU is still available. I don't mind it too much but its still doing it even after I turned deactivated the alerts about a week ago. I just removed her from my favourites as well so I should know tomorrow if it worked. I'm going to be a a loss for words if I ever buy her and still get the emails. It's like MANDARAKE is pleading me to buy her.
    I finally getting around founding a way to Tune Rin to fit my tuning style but I feel a little mixed about the result. She sounds too much like Len :cautious: It's a little choppy since I'm still trying to figure out her phoneme behaviour.
    I just bought the 18" Pop Pikachu today. He is MASSIVE. If I hadn't see him in person i wouldn't have thought of buying him especially not off amazon or ebay. Some listings have the shipping price as expensive as he is. Pictures don't do his size justice.
    I was showing Pictures of my figures to my Aunt while were working and she notes "You don't have a lot of figures of men". I'm just here smiling under my mask because she cracked the code.
    My mom has been telling me that she really wants to sew my DD miku a new dress. She keeps saying the original V2 outfit makes her looks really plain compared to my Miku Figures. :LOL: (DDMIKU is in the same shelving unit as my "miku with you 2018" and "Mebae Miku" [she's to big to put with the others with the V2-V3-V4 outfits]so I sort of get what she means.) If My mom ever finds the time to make a custom outfit for Miku I will post it about it for sure.
    hamano
    hamano
    I wish I could sew clothes. Would be neat to design just the kind of costume I want for my dollfie.
    mobius017
    mobius017
    It's really really nice that your mom supports your hobbies enough that she wants to make clothes for your Miku!
    PSA! I found out through trail and error that you can get Miku english (and maybe also other VB) to sing an American 'I' sound from words like wind, it, and in (not sure what to call it) with Phoneme symbols like 'I@ I' or ' e { ' or 'I {' and splicing the beginning of the note 1/64 (depending on the note and word length).This forces the program to pronounce the phonemes at an accelerated speed. Because of how well Miku's vowels blend it's almost seamless. Different notes react differently to different combination so, it's a bit of a mix and match type of situation. Her default Phonemes are a mess but at least it's possible to make new sounds with what she has
    I finally got accesses to my copy I'm Rin and Len's english VB and I'm testing them now for the first time! I'm pleasantly surprised by their range. I didn't know they would be more audible in a lower octive than Miku. One big problem is that I can't recognize Rin with my tuning style as it is now. I'll need to find some workarounds for that. When I put her voice in a Vsqx made for Miku I can barely even tell that I've changed the voice except for the fact that Rin is slightly more breathy than Miku and that her flow with combination phonemes is a little choppy but thats to be expected since no two VB have the same sounds assigned to the same phoneme symbol.
    So, I just bought Rin\LenV4X from big fish audio this morning at 7:30. (Where I also bought Miku, KAITO, MEIKO and Luka) They've sent me the Download email but I haven't received my serial Number... its almost 2:00. I've sent an email to notify them about it at 9:00. so, I'm losing my mind because I don't want to have to buy them again because of an error that I'm not at fault for.
  • Loading…
  • Loading…
  • Loading…
  • Loading…
  • Loading…