Why Tuning is so Important for Vocal Synthesizers【Kasane Teto AI lite】

Sdílet
Vložit
  • čas přidán 22. 08. 2024
  • this is the same cover I did of Dororo No Uta with Kasane Teto, but with her AI voice.
    If you're new to playing with vocal synths, you might be confused as to what tuning is. To put it simply, it is mainly to do with editing the pitchbends of the keys that your virtual singer is singing in, also known as "training" in japanese. It can also refer to the other parameters you can edit depending on your software. It is not only a way to make your singer sound more realistic, but it is also to give expression that cannot be expressed from the flat keys.
    Here, I tried my best copying the voice delivery of the original singer from the song I'm covering, and as you can hear the base midi sounds so wrong compared to the part with manual tuning.
    It's a misconception that an advanced AI synthesizer like SynthV can give you an instant perfect result with it's automatic AI tuning. Ultimately, the tuning skill of the producer is responsible for delivering an expressive singing voice.
    to be clear, this demonstration is not using SynthV's automatic tuning for the "no tuning" part, but either way, it was pretty bad if not worse than when i removed it.

Komentáře • 25

  • @r.ecollection
    @r.ecollection Před rokem +52

    mixing is also important !!

    • @toiletP
      @toiletP  Před rokem +8

      Absolutely, and it's actually what i am working really hard to learn right now.

    • @r.ecollection
      @r.ecollection Před rokem +1

      @@toiletP us too

    • @ChimeraX0401
      @ChimeraX0401 Před rokem +6

      In Teto's utau voice bank mixing really helps in removing some of the robotness in her voice, that is if you're willing to spend hours of work to make that happen. Though at least in synth v the mixing will be much lighter since the output is much cleaner than utau...

  • @ichi-go7046
    @ichi-go7046 Před rokem +48

    Isn’t this the opening from the old Dororo series? It’s pretty interesting seeing how Teto suits this song!

    • @toiletP
      @toiletP  Před rokem +6

      yes! i just uploaded the full cover too.

  • @Muffin13.mp4
    @Muffin13.mp4 Před 4 měsíci +13

    no tuning sounds like when you put mangio crepe on ai models

  • @littlecherryartist
    @littlecherryartist Před rokem +18

    Do you have any tips on tuning? I'm new and don't know how to tune well

    • @toiletP
      @toiletP  Před rokem +18

      hmm well, my best advice is to listen to the original vocals of a song and tune it based on that. you can also record yourself singing and using that as a reference too only if you're confident enough. just like with drawing you also need a reference for tuning a vocal synth

  • @MothPlum
    @MothPlum Před rokem +23

    I would like the try synth v again but I don’t understand how the changing of the pitches work. How did you learn?
    Also is teto ai only in Japanese? Or did you find and English version
    Thanks! This sounds great btw. Her voice suits this a lot.

    • @mousedarkarts
      @mousedarkarts Před rokem +8

      (let me know if im not supposed to answer, i just love synthV) each voicebank has a "language" option to kind of get it to sing more in the selected language, but every tuner does it differently.

    • @MothPlum
      @MothPlum Před rokem +3

      @@mousedarkarts oh you’re fine. Thank you for replying. And I see! I will look into getting her. And yeah I know everyone does it differently I guess I just would like a guide for what generally one would do to make it sound more natural. Every time I try it’s just random haha.
      I don’t have much experienced with music though. Will see if I can find a tutorial. If you know any good ones I’d appreciate the info!

    • @mousedarkarts
      @mousedarkarts Před rokem +4

      @@MothPlum tbh a lot of UTAU tutorials have tips for doing pitchbends! It takes a lot of practice in synth v though since it's a bit different, but the idea is the same

    • @toiletP
      @toiletP  Před rokem +12

      To change the pitchbends, select the pencil icon and the squiggly lines and you can draw on the piano roll freehand.
      About the language, its complicated. All synthV voicebanks as of now are able to speak chinese, japanese, and english through the cross lingual synthesis, however since this is the free "lite" voicebank, it doesn't have this feature and can only speak Japanese. However, if you do type in english lyrics, synthV will still try it's best to fit the phonemes into something that sounds like english. It works and i have tried it before

    • @Leah_Lusdoc
      @Leah_Lusdoc Před měsícem

      Abt ur Teto Kasane question. She has an English version (On paid vb) and Chinese and Cantonese too

  • @jaquelinepimentel1533
    @jaquelinepimentel1533 Před 3 měsíci +3

    Can I use multilingual on Kasane Teto Lite

    • @toiletP
      @toiletP  Před 3 měsíci

      no sadly, xls is for the paid version only

  • @mireip
    @mireip Před 17 dny

    did u use synth v pro or basic? and what differs between the two?

    • @toiletP
      @toiletP  Před 17 dny

      basic
      basic is free and doesn't have xls, also basic can only use lite voicebanks

    • @mireip
      @mireip Před 14 dny

      ​@@toiletPare lite voicebanks less expensive? thanks for answering tho

    • @toiletP
      @toiletP  Před 11 dny

      @@mireip lite voicebanks are free

  • @zonmbz
    @zonmbz Před 2 měsíci

    how do you remove the tuning

    • @toiletP
      @toiletP  Před 2 měsíci

      you can't
      lol

    • @toiletP
      @toiletP  Před 2 měsíci

      well wait no you can but like it requires you to start over i think
      idk i forgot

    • @Senko_P
      @Senko_P Před 20 dny

      i might be a little late but you gotta select one note, press CTRL + A, then go up to modify (I think) and there should be reset pitch (this resets every note)