• 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.
YOYo_MAMA
Reaction score
958

Joined
Last seen

Profile posts Latest activity Postings About

  • Myth or Fact? Please confirm cause now I'm seeing JP twitter post saying Kizuna AI will have a Cevio AI. SEKAI was just set for reservation and Cevio Pro was just released, can the Dev Team get a breather?
    Exemplar
    Exemplar
    Techno Speech confirmed it

    Plus IA is going to participate in Kizuna Ai's big retirement event too

    Kamitsubaki just made a Youtube channel for SEKAI. Nothing's there yet, but it confirms her Cevio AI release, wonder who will they get for the demo songs?
    Do Google translate and Microsoft Azure T2S count as Vocal Synth?
    Nokone Miku
    Nokone Miku
    Just my two cents, but "vocal synth" usually refers to a synthesized vocalist (i.e. singer). If it isn't musical in nature you would usually refer to it as a "voice synth" or "speech synth." I might argue that "vocal" is singing, "speech" is talking, and "voice" could be either.

    If the program doesn't allow for tuning or adjustment to the output, that is to say, it just lets you put in text and doesn't let you tweak the output, it would usually just be referred to as a "text-to-speech" program rather than a full-on "speech synthesis" program. To me "speech synthesis" would imply more control over the output than your typical text-to-speech program/app.
    YOYo_MAMA
    YOYo_MAMA
    It's still, in it's most basic, a recorded library of vocals that speaks what ever to type in it.
    Vocals, either sung or spoken as both are related to a human voice, are recorded and synthesized to a computer with a purpose. But it can be specialized depending on it's intended target audience.
    This is where the two branches off. One for entertainment/music industry will be different from one for businessman or broke students who just want to know who to speak another language.
    A businessman don't care about the different tones, they might prefer one that can speak 12 or more different language fluently and correctly. Microsoft Jenny is still the most realistic multi-language speech synthesizer I heard, and she speaks 11 languages. It's only on a single tone.
    While a producer would want more than 12 different tones/pitch/speed, but the language may be limited to only 1 or 3. Unless they get a voice provider or the technology to make one sing 12 languages fluently, without that accent. I can still hear a little of that accent in Synth V, but "Engrish" can be accepted by fans so that don't matter.
    But it does it's intended job, you type the words, then it speaks the words.
    For Vocaloid, Cevio, Synth V, Coefont etc, it's the same thing at it's core, You type the words and it speaks/sings the words, but very specialized. With features to change tiny details that some newbies will have idea about, like the tone/pitch/brightness, growl etc. For those who want far, far more creative control over the output than a single bland tone.
    Nokone Miku
    Nokone Miku
    If someone say's they're a "vocalist," a "vocal artist," or that they provide "vocals" I'd usually assume they're referring to singing. If they do "narration," "oration," or "voice acting" I'd assume they're talking about spoken lines of dialogue (or a monologue). And if something is related to "speech" we don't normally associate it with "song."

    That's why I would say that "voice synth" is the broad term while "vocal synth" and "speech synth" are the more specific terms. You see it in the titles of programs like "Vocaloid" and "Voiceroid." Or how people use the term "Talkaloid" to refer to spoken dialog productions that use Vocaloid. You can split hairs all day about technical etymological definitions of words. What matters is the terminology that is useful for quick and clear everyday communication of ideas.

    If you said, "check out my 'vocal synth!'" and held up a Speak'n'Spell I would be really confused. ("Oh, uh... yes, I suppose the Speak'n'Spell does reproduce the sounds of human vocalization from a library of samples, but that is not what I thought you were talking about.")
    I'm not that into KAFU, maybe it's cause I not really a fan on her VB's style, except for just one song, everything else is... meh.
    But it's refreshing to me to see JIN back in the vocal synth business with Chinozo.
    Just recently heard a producer re-released an IA song he made using IA_ai. It's obvious the Cevio AI Version sound better than the Vocaloid Version. Maybe it's cause the vocaloid version sounds like an Orangestar-tuning copycat already.
    ___
    ___
    Can't really be a tuning copycat when the song isn't even tuned nor mixed for that matter. 😅
    Is there a thread completely dedicated to posts like these? Nothing serious just people having fun cause I'm sick of this fickle drama on the community right now.
    Call me insane but I'm seeing weird parallels to 1st Place direction with IA, ONE and Hippy being Aliens who are sent to earth to help it, and David Bowie's Ziggy Stardust, his on-stage persona for his album, who is an Alien meant to help earth, but fails when he let his fame went to his head. Plus David Bowie is also famous for his glam-rock themed outfits.
    Either this is completely coincidental since Aliens are a common trope, or 1st Place producers are fans of British rock and David Bowie. I bet on my non-existed vocaloid collection that next year they will announce the Deer guy as space anime John Lennon and Lion guy as space anime Paul McCartney
    After experiencing the impatience of Game releases and rushed results, I would still rather wait for another 2 years than a rushed piece full of glitches and bugs.

    Currently I'm practicing my patience on the Hollow Knight "sequel" Silksong. Love the first game to bits.
    Found this hidden gem of a Cevio-user in twitter and I am super impressed of how he uses ONE. Don't let the MS paint art fool you this guy have skills. He is on par with the guy who made the Roku and Unknown mother goose ONE covers that was made to a MMD Video by Kreifish.
    AHS is really working overtime with the AI banks cause Maki have both Synthesizer V AND Cevio AI banks now... In the Japanese And ENGLISH language. She's bisoftware and bilingual now.
    Let's see, there's Rikka in AI,Kiritan in AI.
    What's next? Akari in AI?
    Yukari's Cevio design was revealed. And it's kind of a mix bag, like it's trying to be formal and elegant with the dress, shawl and gloves, but later took a turn to try and be spunky & modern, with the headphones and dark color pallete. It's just my thoughts but it may look better if the pallete is more lighter with small black accents, not making black the main color.
    At least the new demo is more refined than nice breathes compared to what they released before, and the Vidol is showing alot of opportunities.
    • Like
    Reactions: ___
    ___
    ___
    Wish theyd try keeping the more casual look her Voiceroid/Vocaloid has, I'd argue that's what made her design initially so appealing.
    After listening to some of IA"s "pray for real" songs. 1st Place is really try cater to the mainstream western genre, e.g. hiphop, modern, rap and mixing english with japanese, than the usual vocaloid songs. it does not help that the composers and lyricist are "professionals" outside the vocaloid community who don't really use the standard vocaloid song style.
    I don't hate it, modern hiphop style is common in other songs like Giga, his work Ready Steady is using a K-pop like beat.
    I like it well enough not to complain about it, The direction will take some getting use to since I'm so used to vocaloid songs being it's own niche outside the mainstream, it's practically it's own genre. The songs can be sung by another human singer like Reol and I won't be surprised, but a vocaloid will have me taken back. I'll admit there are some ranges that can only work well by a human than a vocal synthesizer. I like "DIAMANTE" song the most since it show's IA's range in different style, soft japanese, singing in english and rapping, that one high note is annoying but then I remember I listen to Rolling girl, who have a higher range. I really wish that they credit the tuner, cause making IA rap in english smoothly and well enough that you can understand the words is very difficult and takes so much skill.


    There are complains that this direction of music is butchering IA's brand image, like trying to pander to the mainstream audience than trying to be consistent with IA"s mysterious aura.
    DefiantKitsune
    DefiantKitsune
    It doesn't help that
    a) the rap sounds like it comes from a different song than the rest of the song
    b) it's hilariously obvious the banks are different.
    cafenurse
    cafenurse
    i cant get over songs that use both banks because theyre worlds different,,,, i hate the contrast SO much. im really not a fan of this voicebanks being split up on different engines thing
    Cevio AI Yukari demo was recently released, Now to wait for vflower, (cflower??) IA & ONE demos...

    Cmon Cevio, 1st place & Gynoid, don't make me wait for this pandemic to end before you release something
  • Loading…
  • Loading…
  • Loading…