• 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.
poser
Reaction score
438

Joined
Last seen

Profile posts Latest activity Postings Resources Usernames About

  • i wonder who the non-derivative(? is this the right word) face of synth v might be? im not counting teto or kafu, famous as they are, cuz theyre more popular on their respective engines. maybe rika?
    haha i ended up making a midi of that newest haru demo song...but the producer of the song ended up releasing an official midi of the vocals himself hours after i finished T_T waah
    im not too sure how ai stuff works, would a screamo bank for ai be possible or is it more suited for concatenative synthesis
    IO+
    IO+
    I don't think it's possible, but i like the idea. Problem is AI doesn't know where,when to shout or scream.
    WyndReed
    WyndReed
    Would making a Synth V Vocal Mode for screamo work?
    luwo
    luwo
    I believe it'd be possible, they'd just have to solely be trained on scream/growl vocals so there's no clean vocals interfering. It could definitely lead to some interesting and diverse vocal modes since there isn't just one type of scream.

    I'm really not a fan of how harsh vocals sound when being resampled in concatenative programs though (they feel more prone to looping issues and unwanted artifacts, I often just prefer to use their samples), so I hope we get one at some point.
    as much as i love both synths i feel like cevio kills synth v in terms of having better vocal diversity...idk alot of synth v voices except a few sound samey to me
    IO+
    IO+
    In my opinion it is depend on user input as well, you might aware that people hit AI button and get instant result and the same sound you mention about might come from that, because it's lack of user input, AI end up generate similar singing style and if you listened enough of them you feel that it's sound the same.
    AddictiveCUL (Add)
    AddictiveCUL (Add)
    I kinda agree with both comments. People think just letting AI do the job enough and it usually goes pretty bad, but I agree that CeVio has far way more diverse voices than SynthV. Rose, Ci Flower, Chis-A and Kafu are so spectacular.

    Also, the new voicebank that is about to going out seems so cool! I love it!

    That's a criticism I have about SynthV. Their characters don't sell as much to me as others synths. I can totally buy CUL's ninja fiction and vibe with her distinguishable and beautiful voice, but anything else than Teto, Solaria and Kevin sounds pretty okay to me in SynthV (And, ohh, don't make me start talking about Rose. That girl is fantastic!)
    i hate tuning in cevio - which other synth has a consitent/good pitch exporting results from Utaformatix? I know SV -> CevioAI is...a bit bad
    ___
    ___
    I used to get pretty clean results from UTAU, but not anymore, I think your best bet is tuning in whatever you prefer and then cleaning it in Cevio
    is it ethical to use AI generated lyrics?
    • Like
    Reactions: AddictiveCUL (Add)
    lIlI
    lIlI
    Some AI text generators use writing without the authors' consent, making them unethical. They've been found sampling fanfiction, song lyrics sites, and so on.
    Tortoiseshel
    Tortoiseshel
    I think the only person who can answer this question for yourself is you. Taking in large amounts of text is the way these large language models learn how to do what they do, and in almost all cases that does include copyrighted material. It's up to you whether or not that's unethical. Otherwise, if you have the technical know-how and the processing power, you could try training your own language algorithm on material you're comfortable sampling, like your own writing or public domain/creative commons stuff.
    poser
    poser
    thanks for the input guys :)! ill just keep trying to write my own lyrics for now
    does anyone have ruko's death voice vb?the link seems down or im dumb
    • Like
    Reactions: Ibis
    ___
    ___
    seems to be working for me! the link is up, but if you'll continue having issues with downgliding it I can share it with you.
    poser
    poser
    ___
    ___
    don't sweat it!
  • Loading…
  • Loading…
  • Loading…
  • Loading…
  • Loading…