Clone your voice | How to install Real-Time Voice Cloning toolbox Python? Full tutorial | 2022

Поділитися
Вставка
  • Опубліковано 9 лис 2024

КОМЕНТАРІ • 243

  • @HeyLetsLearnSomething
    @HeyLetsLearnSomething  Рік тому +3

    Build pretrained models to clone your voice using Real-Time Voice Cloning: ua-cam.com/video/qxLF-RegLWs/v-deo.html

  • @HeyLetsLearnSomething
    @HeyLetsLearnSomething  Рік тому +14

    For those who are getting this error "load() takes 1 positional argument but 2 were given" Please try installing librosa "pip install librosa==0.9.2" if not 0.8.1 version!

    • @Bollywood_Songss
      @Bollywood_Songss Рік тому

      I have get this problem....but your suggestion is work...now done

    • @sgquotesmotivation
      @sgquotesmotivation Рік тому

      good solve bro thank you i have much confused why not worked, step 1. try to uninstal librosa by command : pip uninstall librosa, n then try to install pip install librosa==0.9.2

    • @LiveStreamLemo
      @LiveStreamLemo Рік тому

      thank you you save me

  • @TMGreycoat
    @TMGreycoat Рік тому +15

    This is such an excellent tutorial. As someone who is fascinated with this but doesn't have the technical knowledge to use it, your detailed instructions really made a big difference.

    • @kustomize93
      @kustomize93 Рік тому +3

      this comment made me watch your only video and it was a great decision I made in life

    • @IKnowThatMyRedeemerLiveth
      @IKnowThatMyRedeemerLiveth Рік тому

      @@kustomize93 this comment made me watch @TheOneWhoIsTheTwo's only video and it was a great decision I made in life

  • @JaveGeddes
    @JaveGeddes 2 роки тому +13

    I remember seeing that the pre-trained model link was down. We definitely need a tutorial for making our own models, & using more voice clips to make it better. Having an option that doesn't require an insane video card would also be great.

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +3

      HI. Have you checked this?
      Build pretrained models to clone your voice using Real-Time Voice Cloning: ua-cam.com/video/qxLF-RegLWs/v-deo.html

  • @surajitpatro
    @surajitpatro Рік тому +11

    Hi, How to fix this? AttributeError: module 'numpy' has no attribute 'float'. Did you mean: 'cfloat'? I tried installing numpy 1.20 too

    • @PRIMEDECODER
      @PRIMEDECODER 22 дні тому

      did you get the solution of 'numpy has no attribute float'?

  • @TUITSAMYUKTHAK
    @TUITSAMYUKTHAK 8 місяців тому

    It worked just fine. Your explanation is on point. Thankssssss a ton!

  • @pallathukalam
    @pallathukalam Рік тому

    I know little about Python and I spent 5 days to figure out how to install different voice cloning programmes from different people but failed miserably. But you are a great teacher. This is the only video that helped me to complete the task without getting even a single error. I must thank you for the way you explained with great clarity and I never skipped the ads on your video as that is the only way I can show my gratitude to you for this excellent video. Thank you again. Deva (Kerala, India)

    • @ashishvalentine2979
      @ashishvalentine2979 Рік тому

      Hey I want to connect with you I have a project and a proposal

  • @itswafflebob
    @itswafflebob Рік тому +5

    Great tutorial! Thanks a ton for putting this together.
    One thing: I noticed when copying the code from your site that the line, "conda create -name voice-clone python=3.9" has a typo in it which prevents the command from executing properly. Just adding the extra hyphen before name solved the problem if you'd like to update that on your site.

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +2

      Thanks, Bob. We will check that and make the necessary updates :)

    • @NathanChambers
      @NathanChambers 11 місяців тому +1

      @@HeyLetsLearnSomething 7 months later and you still haven't fixed the mistake. At least it's a simple fix if you just read the error output :P

  • @danielcardone8622
    @danielcardone8622 2 роки тому +24

    Can someone please make a video detailing how to build a pre trained model from scratch to use with this tool. I know it involves downloading audio, chopping it up, transcribing it but what I don't know is how to format and organize it all to use with this tool. Does anyone teach that part clearly? Alot of these videos say "download the pretrained model, or use your own" I want to build my own. But I can find a clear video that doesn't create more questions than answers. Advice please.

    • @EyeBallGamers
      @EyeBallGamers Рік тому +4

      Hi did you find anything?

    • @someonesomewhere8869
      @someonesomewhere8869 Рік тому +2

      Hi did you find anything?

    • @adamlamay5846
      @adamlamay5846 Рік тому +2

      Any one got any answers or suggestions to drop?

    • @Blackkspot
      @Blackkspot Рік тому +2

      Hello. Does building a pretrained model mean that you could make this work with other languages?

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +4

      Hi Daniel, pretty late but here is a guide on how you can build pre-trained models to clone your voice using Real-Time Voice Cloning: ua-cam.com/video/qxLF-RegLWs/v-deo.html

  • @easyjapaneseforall
    @easyjapaneseforall Рік тому +4

    wow man, it worked smoooothly, thank you so much, you should make tutorials about things, you explain very detail. Thanks Again

    • @dracus3402
      @dracus3402 Рік тому

      I couldn't do.Can you help me?

    • @easyjapaneseforall
      @easyjapaneseforall Рік тому

      @@dracus3402 if you can add the minute where you got an error or got confused I can have an idea or someone else can help you, so get in the time you got stuck and we can figure it out. Or maybe your console message.

  • @ivanfilatov4414
    @ivanfilatov4414 2 роки тому +2

    Thanks for the video, Pal! It would be great to see more videos related to deep fake and voice cloning on your channel .

  • @Max-ei2so
    @Max-ei2so Рік тому

    Thank you for your video I really needed a how-to-install guide, otherwise it would have been too complicated and exhausting for me to try out this exciting project. 😄

  • @KimJaejoong-rm3zq
    @KimJaejoong-rm3zq Рік тому +2

    Please am constantly getting an error saying “Attribute error: module ‘numpy’ has no attribute ‘float’

  • @danielcardone8622
    @danielcardone8622 2 роки тому

    Dude!!! I always wanted thus tool but never could get it right until this video. THANKS!!!

  • @samturner64
    @samturner64 Рік тому

    Thanks for the tutorial, very easy to follow even if not using the dataset library! I recommend this video for any problems with the setup!!

  • @SidDaysKaraoke
    @SidDaysKaraoke 2 роки тому +1

    Thank you so much! I had to modify some steps slightly, but was able to get it to work using my RTX 3080 GPU.

  • @rdn404
    @rdn404 8 місяців тому

    Thanks for the lesson! It worked perfectly

  • @PRIMEDECODER
    @PRIMEDECODER 22 дні тому

    how to fix 'numpy has no atrribute float'?

  • @redthunder6183
    @redthunder6183 Рік тому +1

    set the speed to 1.25x so that its not painfully slow

  • @Sponta_
    @Sponta_ 10 місяців тому +2

    I did everything including loading the dataset, but just when I want to use the voice, when I press Load, I get this error:
    TypeError: mel() takes 0 positional arguments but 2 positional arguments (and 3 keyword-only arguments) were given
    I'm stuck, do I have to reinstall everything again?

    • @ankidokolo
      @ankidokolo 7 місяців тому

      Hi. I have been working on this for 2 days now. Did you find any solution for it. If you could kindly share it I would very very... very much appreciate it.

    • @Sponta_
      @Sponta_ 7 місяців тому +1

      @@ankidokolo i asked to bing ai, i just remember i had to manually search in some text files and change some specific deprecated parts of the code

    • @walfischrobbe
      @walfischrobbe 6 місяців тому

      @@ankidokolo downgread your librosa version. pip install "librosa==0.9.1"

  • @timtomnec
    @timtomnec 6 місяців тому

    A few things
    to change from the C drive type the driver letter then colon F: (Cd will only change the directory you are in aka folder it does not do any thing to the drive)
    don't save stuff to the desktop its more work then it saves
    up arrow will automatically retype the last line, and Tab will do an auto complete most of the times

  • @aznetvietnam
    @aznetvietnam Місяць тому

    Hi, I got an error when I run the demo: "TypeError: melspectrogram() takes 0 positional arguments but gave 2 positional arguments (and 2 keyword-only arguments)"
    How can I fix it? thanks

  • @negi04
    @negi04 Рік тому

    THANK YOU FOR THE EASY TUTORIAL :)

  • @dhanielbawias9933
    @dhanielbawias9933 Рік тому

    Thanks for the tutorials. I have a few questions : what method or algorithm do you use bro????

  • @VaibhavShewale
    @VaibhavShewale Рік тому +2

    well i got "ImportError: DLL load failed while importing _imaging: The specified module could not be found." and dont kno wha is the issue, i even install matploit and numpy but still
    update -
    been trying to do this from past 5 days no solution or progress been stuck, time to abandon it!

    • @omaremad375
      @omaremad375 2 місяці тому

      install the windows 10 sdk mentioned in the video and windows 11 SDK plus uninstall pillow and reinstall it again but with specifying the version ,it depend on your python version , if your python 3.11 ,you can just install pillow >=9.3 if your python version 3.9 ,you can install pillow >=8.3.2 but you can not install pillow 8.3.1 , you need to uninstall it first or move it to another directory
      that are the step that worked for me

  • @Show_Bits
    @Show_Bits Рік тому +1

    Thanks. I get to the very end though, it says all requirements are met and then when I run the demo... nothing happens? it doesn't open the toolbox. Any suggestions?

    • @rachitmehta2634
      @rachitmehta2634 Рік тому

      Even I got to the very end and when i ran the demo nothing happened, I think i read if you have windows 11 you may face this error

    • @Show_Bits
      @Show_Bits Рік тому

      @@rachitmehta2634 Actually I got it to work. I hadn't deleted the files that were already in there upon install. There were random jpegs and stuff with labels like "frame_003" etc that I deleted. once the folder was empty it worked fine.

    • @mathislabory6753
      @mathislabory6753 Рік тому

      I have the same issue. Any updates ?

    • @sgquotesmotivation
      @sgquotesmotivation Рік тому

      exactly, sure i use windows 11 n doesnt open, any solutions ?

    • @ThinkPsycho
      @ThinkPsycho 10 місяців тому

      i have the same issue

  • @utf-238e8
    @utf-238e8 Рік тому

    You are a savior, thank you for your video

  • @browntown7663
    @browntown7663 Рік тому +3

    ModuleNotFoundError: No module named 'numpy"
    tried pip install still not working, please help

  • @doraemon_16
    @doraemon_16 Рік тому +3

    melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given. ERROR. ¡¡HELP ME PLEASE!

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +2

      Can you please try installing this version of librosa. Thanks.
      pip install librosa==0.9.2

    • @CriptoBroh
      @CriptoBroh Рік тому

      @@HeyLetsLearnSomething worked! Thanks a lot!!

  • @Bigjuergo
    @Bigjuergo Рік тому +1

    what do i do wrong if my voice does not sound like me?
    do i need longer training samples of my voice? or do i need more training samples form my voice?

  • @chase_parca
    @chase_parca Рік тому +1

    I have the error exception: load() takes 1 positional argument but 2 were given how do I fix it?

    • @pawelarz
      @pawelarz Рік тому

      I have such a problem too. did you manage to fix the error?

  • @HeyLetsLearnSomething
    @HeyLetsLearnSomething  2 роки тому +5

    Install and Create Deepfakes (first order model - python) on Windows: ua-cam.com/video/foGYFroCPn0/v-deo.html

    • @itagakixxx
      @itagakixxx Рік тому +1

      why is this called Real-Time Voice cloning but the synthesize and vocode process i pretty slow ? and i want to ask "how do you think AtheneWins make his AI Streamers ?

    • @akissot1402
      @akissot1402 Рік тому

      C:\Windows\Desktop ? thats windows 98

    • @BlvnkDriverPhonkR32
      @BlvnkDriverPhonkR32 Рік тому

      @@itagakixxx with a gpu deadass 💀💀💀💀

  • @adrianlets
    @adrianlets Рік тому

    Excellent video, but it works only for the English language, is there a way for other languages. Greetings

  • @arbozfermozshodan70
    @arbozfermozshodan70 2 роки тому

    Hello bro
    Behat behat Dhanvād brātar thyöv sirf matrar shaqs sin tat ber main taqneek e haqyie dvedanat
    Love from India

  • @dethmate3395
    @dethmate3395 Рік тому +2

    Could you help me with this error?
    TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +5

      Can you please try installing this version of librosa. Thanks.
      pip install librosa==0.9.2

    • @dethmate3395
      @dethmate3395 Рік тому

      @@HeyLetsLearnSomething will do. Thanks!

    • @aicodehub
      @aicodehub Рік тому +1

      @@HeyLetsLearnSomething thanks it worked :)

    • @powerThawtz23
      @powerThawtz23 Рік тому +1

      @@HeyLetsLearnSomething I tried reinstalling and it worked! You are such an amazing instructor. This was my first attempt working with Python and you gave me the confidence through your solid and detailed explanations to successfully install this. I can't thank you enough!

    • @whatevermhen8824
      @whatevermhen8824 Рік тому

      @@HeyLetsLearnSomething please help im having this error during launching
      Error in sys.excepthook:
      Original exception was:

  • @asbhd526
    @asbhd526 Рік тому +4

    I have this error: TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given does anybody know how to resolve it?

    • @daldanus
      @daldanus Рік тому

      I also have this. No idea how to solve it.

    • @ickaruus4909
      @ickaruus4909 Рік тому

      Found an answer! Type this: pip install librosa==0.9.2
      After that it worked for me!

    • @Satwr
      @Satwr Рік тому

      @@ickaruus4909 thx!! worked for me!

    • @whitebird77
      @whitebird77 10 місяців тому

      ​@@ickaruus4909yes , it's work 😊 , thanks 🙏👍🏻

  • @भूमिकर्माधिकार

    What would you suggest for Androids users? Own voice TTS creation rltd?

  • @ConhecimentoArtificial-lo7ti

    hello does it work only for english or for other languages too?

  • @Dog_Daddy
    @Dog_Daddy Рік тому

    Thank you very much, but how do i download SYNTHESIZER manually?

  • @josu-5565
    @josu-5565 2 роки тому +1

    Works with others languages?

  • @walfischrobbe
    @walfischrobbe 6 місяців тому +1

    i alwas get the error... can someone please help me ? i am very new to this stuff, so i dont really know how solve it.
    TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given

    • @walfischrobbe
      @walfischrobbe 6 місяців тому +2

      just found the solution. im gonna post it here, just in case someone else has this problem to. the new librosa version is the problem. downgrade to 0.9.1
      pip install "librosa==0.9.1"

    • @SwayingLeaves
      @SwayingLeaves 4 місяці тому +1

      @@walfischrobbe thanks man

    • @PRIMEDECODER
      @PRIMEDECODER 22 дні тому

      @@walfischrobbe thank u so much

    • @PRIMEDECODER
      @PRIMEDECODER 22 дні тому

      @@walfischrobbe do you have any solution for 'numpy' has no attribute 'float'? i don't know how to solve this?

  • @louisgelasgamer7553
    @louisgelasgamer7553 10 місяців тому

    thank you very much! But but how to do it with French voices ?

  • @courtneyb6154
    @courtneyb6154 Рік тому +1

    just use the entire folder URL such as cd C:\Users\user\Desktop\Voice-Master . No need to chase the entire URL piece by piece.

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Thanks. Yes, the easiest way is to copy the URL and type cd like cd C:\.... and enter. For other videos, we have shown it that way. Thanks.

  • @Krishna-zw9wl
    @Krishna-zw9wl 21 день тому

    Sir I need urgent pls tell me that sv2tts windows it's not coming is there any solution?

  • @danieliosef
    @danieliosef Рік тому

    Thank you Sir for the mega tutorial. It took me a whole afternoon. However, the audio is really bad although the source of the clone audio is perfect. would you suggest something?

  • @Footballlegendary25
    @Footballlegendary25 Рік тому

    I was trying to do this and mistakenly canceled. Now I'm being told it already exist. What do I do?

  • @ashishvalentine2979
    @ashishvalentine2979 Рік тому

    Can we do a 1 hour audio ?

  • @JohnBallard-c1p
    @JohnBallard-c1p Рік тому +1

    how to fix this? pls help me "TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given "

    • @echoleader5733
      @echoleader5733 Рік тому

      same problem

    • @aznetvietnam
      @aznetvietnam Місяць тому

      @@echoleader5733 Can you please try installing this version of librosa. Thanks.
      pip install librosa==0.9.2

  • @MizanurRahman-ox9zd
    @MizanurRahman-ox9zd Рік тому

    can i do it with my own native language,sir?

  • @matthewtull
    @matthewtull Рік тому

    The voice-clone is not on the Anaconda repo. How do I get it now?

  • @checkupcodes
    @checkupcodes Рік тому

    I want this to give answers faster. Does anyone know a way around this. Thanks for the video

  • @selvinsalve
    @selvinsalve Рік тому +3

    The toolbox is not launching in my pc, what to do?

  • @IbrahimTolid
    @IbrahimTolid Місяць тому

    Does this work on Mac?

  • @prod_guzz
    @prod_guzz Рік тому

    thank u SOOO much, i will try to recriate GLaDOS voice

  • @saifpathan3505
    @saifpathan3505 9 місяців тому

    i have done everything correctly but the demo_toolbox is not opening what can be the issue?

  • @kostas9849
    @kostas9849 Рік тому

    Do you know if this works with foreign languages?thank you 🤠

  • @sohailhosseini2266
    @sohailhosseini2266 Рік тому

    Thanks for sharing!

  • @ickaruus4909
    @ickaruus4909 Рік тому

    I have a really weird problem I think. At the very end when trying to import voice examples into the program the command doesn't work. I think this is because the user name for my pc consists of 2 words with a space. Is there another solution to this than changing the name?

  • @LyllianaofMirrah
    @LyllianaofMirrah Рік тому

    the current version of python is 3.10 or i need 3.9?

  • @foundfpvfootage
    @foundfpvfootage Рік тому

    The path traversing using cd was painful to watch, but nice tutorial overall :D

  • @Soviet_Fiction
    @Soviet_Fiction 2 роки тому

    Sir, thank you very much. I got it all

  • @wildingoncam
    @wildingoncam Рік тому

    Thanks.. I need this for my project and I was getting stuck for a second as I'm currently new to conda.

  • @prod_guzz
    @prod_guzz Рік тому

    can i use pip instead of conda?

  • @dermitdemmikro
    @dermitdemmikro Рік тому

    i have this error . cannot load library 'libsndfile.dll': error 0xc1 ?

  • @free_thinker4958
    @free_thinker4958 Рік тому

    hello thanks man , i still have a question !! what is the optimal duration of the audio that we use if we don't use the dataset

    • @FerSpirit
      @FerSpirit Рік тому

      the original paper said 5 sec

  • @ezequielacosta9662
    @ezequielacosta9662 Рік тому

    melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given

  • @sgquotesmotivation
    @sgquotesmotivation Рік тому +1

    demo_toolbox_py doesnt open, why ? anyone know this solutions ?

  • @indraramya795
    @indraramya795 Рік тому +1

    Sv2tts toolbox can not be launched
    No error is shown

    • @nightsky1664
      @nightsky1664 3 місяці тому

      same problem, please let me know if you solved it
      im stuck :(

  • @НастяСироткина-з5с

    hello, could you please tell me if there is an option to integrate translator from english to chinese (ive already got it) to the table of speech playback or it has not ability to read hieroglyphs and need to be applied for g2p?
    thank you

  • @Cryptopromis
    @Cryptopromis Рік тому

    Thanks it worked for me

  • @englishhugo4576
    @englishhugo4576 Рік тому

    If my voice is spanish voice it work?

  • @zack2804
    @zack2804 2 роки тому +1

    How do you train the voice?

  • @_phenomeme_
    @_phenomeme_ Рік тому

    9:15 I believe you could have just copied the Voice folder path after cd and it would skip all the steps straight to the point, there is no need to mess with two dots, or to go to the desired folder step-by-step

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Thanks, that's true. We realized the shortcut later, but we have actually replied with the same information in the comments. Simply typing "cd C:/..." will take you to the folder. Anyway, thanks for pointing it out!

  • @nightsky1664
    @nightsky1664 3 місяці тому

    Didn't work for me :(
    Now how do i uninstall everything i just installed?

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  3 місяці тому +1

      Sad to hear that. To uninstall it you'd need to remove the env you created..this code will remove the env (conda remove --name voice-clone --all) & Uninstall anaconda from cPanel...Thanks.

  • @LyllianaofMirrah
    @LyllianaofMirrah Рік тому

    thx i got it working i need to download trained files

  • @MohamedHelmy-b9o
    @MohamedHelmy-b9o Рік тому

    Does it work with other languages such as Arabic

  • @MIGUELCHUNGA-p6p
    @MIGUELCHUNGA-p6p Рік тому

    I get this error, what should I do? -- "exception: load takes 1 positional argument but 2 were given"

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Please try installing librosa "pip install librosa==0.9.2" if not 0.8.1 version!

  • @YouTubeStalker
    @YouTubeStalker 2 роки тому

    Will this work on windows 8.1?

  • @ntinosaggelopoulos6718
    @ntinosaggelopoulos6718 2 роки тому

    Can the voice be sound in difference languages

  • @RamonValdez2014
    @RamonValdez2014 Рік тому

    Need help here please, I didn't pass the vocoder synth test and got this error "TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given"

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +1

      It probably is because of the Librosa. Please try using the librosa vesrion 0.9.2!
      - Check the Librosa version: conda list librosa
      - Uninstall Librosa: conda uninstall librosa
      Or
      pip uninstall librosa
      - Install librosa 0.9.2: conda install -c conda-forge librosa=0.9.2
      or
      pip install librosa==0.9.2

    • @RamonValdez2014
      @RamonValdez2014 Рік тому

      @@HeyLetsLearnSomething I uninstalled the prior librosa version and installed 0.9.2. but can't test the vocoder synth still, but thanks anyway.

  • @optimusdata57
    @optimusdata57 Рік тому

    Hello I have this problem .. Please can you help me? "melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given"

  • @laikhandigitalmarketer5256
    @laikhandigitalmarketer5256 Рік тому

    Please help me whole 2 days later i have open toolbox but when i click Load button it shows "Exception: load() takes 1 positional argument but 2 were given" help me

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Please try installing 0.9.2 version of librosa. Hope this helps. Thanks.
      pip install librosa==0.9.2

  • @omairahmad4155
    @omairahmad4155 Рік тому

    TypeError: can't convert np.ndarray of type numpy.object_. The only supported types are: float64, float32, float16, complex64, complex128, int64, int32, int16, int8, uint8, and bool.. I am facing this issue in sv2tts please someone help me as soon as possible.

  • @demotest-im9pf
    @demotest-im9pf Рік тому

    after selecting speaker if i click on load button then it shows load() takes 1 positional argument but 2 were given

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Please try installing librosa "pip install librosa==0.9.2" if not 0.8.1 version!

  • @ridetoheaven777
    @ridetoheaven777 Рік тому

    Receiving error when running demo state error in sys.excepthook: please can someone help me to resolve this error

  • @alessandrogargano26
    @alessandrogargano26 Рік тому +1

    please add italian pre trained model

  • @blancatello5451
    @blancatello5451 Рік тому

    there is no LibriSpeech folder

  • @privatesoft5006
    @privatesoft5006 Рік тому

    Dose Support Arabic Voices

  • @ShattaGold
    @ShattaGold Рік тому

    Awesome video

    • @ShattaGold
      @ShattaGold Рік тому

      Please can we get a free voice changing microphone for pc

    • @ShattaGold
      @ShattaGold Рік тому

      Please can we get a free voice changing microphone for pc

  • @sachinbhoi5727
    @sachinbhoi5727 Рік тому

    Hello, I want the trained model of custom voice, I want to make api of it

  • @falitosixgod7302
    @falitosixgod7302 Рік тому

    How can i extract the libri speech

  • @cheekangkoh3321
    @cheekangkoh3321 Рік тому

    Need help here please, I didn't pass the vocoder synth test and got this error "AttributeError: module 'numpy' has no attribute 'float'.
    `np.float` was a deprecated alias for the builtin `float`. To avoid this error in existing code, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here."

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +1

      Can you please check the numpy version installed:
      conda list numpy
      (If the version is different to 1.20.3 then try installing 1.20.3 version):
      pip install numpy==1.20.3
      (If this doesn't install directly, try uninstalling numpy and run the installation code):
      pip uninstall numpy

    • @cheekangkoh3321
      @cheekangkoh3321 Рік тому +1

      @@HeyLetsLearnSomething Thanks a lot, I want to know how to export out the cloned voice?

    • @zjionhongtan8354
      @zjionhongtan8354 Рік тому

      @@HeyLetsLearnSomething
      It shows this after I install numpy 1.20.3, not sure how to solve:
      ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
      matplotlib 3.8.0 requires numpy=1.21, but you have numpy 1.20.3 which is incompatible.
      numba 0.58.1 requires numpy=1.22, but you have numpy 1.20.3 which is incompatible.
      scipy 1.11.3 requires numpy=1.21.6, but you have numpy 1.20.3 which is incompatible.

  • @echoleader5733
    @echoleader5733 Рік тому

    Help please what is? TypeError: CCompiler_spawn() got an unexpected keyword argument 'env'

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Can you please let us know the Python version you have installed? Please make sure you are using version 3.9 and install numpy (1.20.3) pip install numpy==1.20.3
      Thanks.

  • @LyllianaofMirrah
    @LyllianaofMirrah Рік тому

    do a Stable Diffusion installation guide PLS

  • @sanZeeet
    @sanZeeet Рік тому

    yesterday test did not passed but today passed with librosa0.9.2 but today I waste whole day each and every things installed despite having toolbox is not opening.....my luck never support me.....help me... whenever I am giving command for toolbox opening it do not giving any message except the flie directory path

    • @drip4558
      @drip4558 Рік тому

      What is the error message I can help you

  • @TheBestSib
    @TheBestSib Рік тому

    TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given,
    what about that error ?

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому +1

      Please try installing librosa "pip install librosa==0.9.2" if not 0.8.1 version!

    • @TheBestSib
      @TheBestSib Рік тому

      @@HeyLetsLearnSomething it helps thank u so much! :D

    • @TheBestSib
      @TheBestSib Рік тому

      @@HeyLetsLearnSomething One more Ques how can i froce GPU use?

  • @zaneykalip
    @zaneykalip Рік тому

    i go the GPU route and got this error "TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given"

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Please try installing librosa "pip install librosa==0.9.2" if not 0.8.1 version!

    • @zaneykalip
      @zaneykalip Рік тому

      @@HeyLetsLearnSomething error during pytorch install

  • @AkuKucing-wr9gj
    @AkuKucing-wr9gj 4 місяці тому

    only engglish accent sir?

  • @sanZeeet
    @sanZeeet Рік тому

    I am facing a problem just ahead of passing the test otherwise it ready to use Message says- Load() 1 positional argument but 2 were given
    Error: melspectrogram
    How to resolve....thank you for the whole tutorial I am watching more n more with full ads to support you

    • @HeyLetsLearnSomething
      @HeyLetsLearnSomething  Рік тому

      Thanks, Sanajit. Please try installing this version of librosa: pip install librosa==0.9.2

    • @sanZeeet
      @sanZeeet Рік тому

      This time error in Sys.excepthook...
      datasets_root:none,models_dir:saved_models,cpu:False,seed:none

    • @adityabarman2002
      @adityabarman2002 11 місяців тому

      The wav_to_mel_spectrogram function seems to be using the librosa.feature.melspectrogram function incorrectly, leading to the TypeError you mentioned earlier. According to the error message, the melspectrogram function doesn't expect any positional arguments, but it's receiving two.
      The correct usage of librosa.feature.melspectrogram is to provide the audio waveform as the first argument and the sampling rate as the sr parameter. However, it seems that you are passing the waveform as the first argument and the sampling rate as a positional argument.
      Here's how you can modify the wav_to_mel_spectrogram function to fix the issue:
      @@sanZeeet

    • @adityabarman2002
      @adityabarman2002 11 місяців тому

      def wav_to_mel_spectrogram(wav):
      """
      Derives a mel spectrogram ready to be used by the encoder from a preprocessed audio waveform.
      Note: this is not a log-mel spectrogram.
      """
      frames = librosa.feature.melspectrogram(
      y=wav,
      sr=sampling_rate,
      n_fft=int(sampling_rate * mel_window_length / 1000),
      hop_length=int(sampling_rate * mel_window_step / 1000),
      n_mels=mel_n_channels
      )
      return frames.astype(np.float32).T

  • @optimusdata57
    @optimusdata57 Рік тому

    So turkish or other language ?