Training Tortoise TTS on Other Languages (Japanese)

Sdílet
Vložit
  • čas přidán 6. 09. 2024
  • Links referenced in the video:
    AI Voice Cloning - • Local AI Voice Cloning...
    Hardware for my PC:
    Graphics Card - amzn.to/3pcREux
    CPU - amzn.to/43O66Ir
    Cooler - amzn.to/3p98TwX
    RAM - amzn.to/3NBAsIq
    SSD Storage - amzn.to/42NgMFR
    Power Supply (PSU) - amzn.to/430bIhy
    PC Case - amzn.to/447499T
    Mother Board - amzn.to/3CziMXI
    Alternative prebuilds to my PC:
    Corsair Vengeance i7400 - amzn.to/3p64r22
    MSI MPG Velox - amzn.to/42MnJHl
    Cheapest and PC recommended:
    Cyberpower 3060 - amzn.to/3XjtZoP
    Come join The Learning Journey!
    Discord - / discord
    Github - github.com/Jar...
    TikTok - / jarodsjourney
    If you found anything helpful, please consider supporting me and the content I am trying to produce!
    www.buymeacoff...

Komentáře • 27

  • @3k3k3
    @3k3k3 Před 7 měsíci +2

    The Holy grail. Pretty much no one covers this. Really looking forward to the result.

  • @shovonjamali7854
    @shovonjamali7854 Před 6 měsíci

    Another great one! Were the data you used for this training from a single speaker or did you mix? If I train with a male speaker and later I want to clone a zero-shot female speaker will this model be able to do it? Please help.

  • @PowerRedBullTypology
    @PowerRedBullTypology Před 7 měsíci +1

    Hey Jarod, would you know if there is an AI that can change the accent of voices, especially when the people are singing? There are these voices like vocaloid/synth V but some singers have pretty thick accents when speaking english. I was curious if there is a way to improve that.

    • @Jarods_Journey
      @Jarods_Journey  Před 7 měsíci

      Well, if you use RVC with index set to 1, it sometimes keeps the native accent of the speaker trained on when using it for inference. You might wanna check that out

  • @myte1why
    @myte1why Před 7 měsíci

    wow, good work man

  • @SiddharthTripathi365
    @SiddharthTripathi365 Před 7 měsíci

    Awesomeee. Would love to see how this comes out!

  • @ElmorenohWTF
    @ElmorenohWTF Před 7 měsíci

    Please keep doing this!! I'm interested in using tortoise tts in Spanish in Google Colab and maybe with something like this I can achieve it. Do you think it can work in another language well without the need for a dataset of hundreds of hours? I don't know, maybe using a pre-trained model or something.

    • @Jarods_Journey
      @Jarods_Journey  Před 7 měsíci

      It should work with other languages, BUT, preparation of data and training in that language (Spanish) would need to be done

    • @ElmorenohWTF
      @ElmorenohWTF Před 7 měsíci

      ​@@Jarods_Journey I would do it without a problem, could you make a tutorial to teach how to train tortoise tts in other languages when you can? Thanks for replying

  • @Mika43344
    @Mika43344 Před 7 měsíci

    great video🔆

  • @bomar920
    @bomar920 Před 7 měsíci

    Looking forward to the result .

  • @amedyasar9468
    @amedyasar9468 Před 6 měsíci

    Hi, I am looking for voice swap training. Do you have any suggestion??

  • @mikkuu7747
    @mikkuu7747 Před 7 měsíci

    Is it possible to change already existing ai voices accent with tortoise? If not, do you have any idea where it can be done?

    • @Jarods_Journey
      @Jarods_Journey  Před 7 měsíci

      You'll need to Finetune the tortoise model on the voice/accent, though, idk how effective it'll be. Depends on the amount of data you have

  • @Shadow-Veil
    @Shadow-Veil Před 7 měsíci

    I just heard the song for the first time why does it go so hard lmfao🤣

  • @bhargavk1515
    @bhargavk1515 Před 7 měsíci

    Can you make a low compute audiobook maker. for both windows and mac users please 😭😭😭😭

  • @ramimithalouni6592
    @ramimithalouni6592 Před 7 měsíci

    Interested to see the result. Can you use the tokenizer of xtts?

    • @Jarods_Journey
      @Jarods_Journey  Před 7 měsíci +2

      I had initially thought of this, but tortoise needs a tokenizer with a width of 256 and xtts has a larger one. I may look into adapting the tortoise architecture to see if it can fit larger ones though

    • @ramimithalouni6592
      @ramimithalouni6592 Před 7 měsíci

      @@Jarods_Journey great. I will follow you to see if that works. Do you think training tortoise will get you better result than fine tuning xtts?

  • @TheBestgoku
    @TheBestgoku Před 5 měsíci

    is there a UI for this ?

  • @johnlenoob6951
    @johnlenoob6951 Před 7 měsíci

    Love from Paris

  • @ahmetab06
    @ahmetab06 Před 7 měsíci

    What kind of size data, hardware and hours are needed for a language like German for a quality train? According to your predictions

    • @Jarods_Journey
      @Jarods_Journey  Před 7 měsíci

      Dunno yet, but I will have a better answer to this later.

    • @ahmetab06
      @ahmetab06 Před 7 měsíci

      @@Jarods_Journey we are waiting for u

  • @AntiAnti
    @AntiAnti Před 7 měsíci

    How have you converted Japanese to Latin characters?

  • @sataz101
    @sataz101 Před 7 měsíci

    ❤ thank you👍💯