Clone ANY Voice for FREE Using AI & Your PC

[@Emit.Reviews]

UPDATE: This tutorial doesn't work after Python version upgrade. I've updated the tutorial here kzbin.info/www/bejne/noC2pHqFoql2adk

[@AlejandroVales]

Humble direct and well presented.. Perfect video!

[@WesleyPaulinoCoelho]

Can you help me? I'm having this error when generating the audio: ImportError: cannot import name 'deprecated' from 'typing_extensions' (C:\Users\coell\anaconda3\lib\site-packages\typing_extensions.py)

[@__Pengy]

Im having the same issue. tell me of you find a solution please

[@realitybites606]

Same, any help would be appreciated

[@justarandomguy6247]

same mane

[@newwonderer]

seems to be author of this channel is shitting on us, so no help

[@OskarsBestD]

can anyone help with this? im getting ValueError: 'build\\py3k\\scipy' is not a directory [end of output]

[@wirelessriver7]

same here

[@kaymat2368]

Same too, several minutes trying to install this trash only to giving that error

[@Emit.Reviews]

Check out the pinned comment. Ive made an update to this method

[@DeadEyeZJhinBRUH]

what to do to speed up the Processing every commands. 1 text to speech = takes 15 minutes to finish. and till I can see the results

[@clyphx]

Me after installing everything and realizes that I don't have an NVIDIA GPU

[@GnA_000]

Me after installing everything and my Nvidia gpu runs out of memory somehow when running tenslorflow models.

[@dustinrolstad752]

I get right up until the part where you need to install Git, and then the 'next' button in the installation wizard is non-responsive. Love it when that happens.

[@Emit.Reviews]

[APRIL 26th UPDATE IF IT DOESNT WORK FOR YOU] I've released an updated video that showcases a new installation method that should work - kzbin.info/www/bejne/noC2pHqFoql2adk ✅ Hostinger $1.79 ➜ emit.reviews/go/hostinger-2023?ai CODE for extra 10% OFF ➜ EMITREVIEWS _____________________________________________ [Old info] To install Tortoise-TTS use: git clone github.com/neonbjb/tortoise-tts.git cd tortoise-tts python -m pip install -r ./requirements.txt python setup.py install You can paste it all at once, not line by line.

[@Artificial_]

or you could just link the colab version

[@thisisashan]

I'm getting errors with llvmlite "ModuleNotFoundError: No module named 'llvmlite'" When I tried to install llvm I got another error that it was version 11 and it needs version 14. New to this stuff. help?

[@Emit.Reviews]

Watch my 2nd video. I show how to fix the errors. I have linked it in the comment above

[@thisisashan]

@@Emit.Reviews ty ty will do. sorry just now found this first video :)

[@thisisashan]

@@Emit.Reviews Oh... and thank you while I have your attention :) Your second video did indeed clear everything up. wondering if there is a webapp for tortoise yet.

[@CamiloMonsalve]

I'm having this error: FileNotFoundError: Could not find module 'C:\Users\PC\.conda\envs\tortoise\Lib\site-packages\torchaudio\lib\libtorchaudio.pyd' (or one of its dependencies). Try using the full path with constructor syntax. (tortoise) PS D:\tortoise-tts\tortoise-tts>

[@serikazero128]

Hmm, my only issue with Eleven labs is that its very restrictive in languages. I'm looking for an option that can do this for any langauge

[@garzsom]

Me too... maybe one day

[@volatile2005]

RuntimeError: PytorchStreamReader failed reading zip archive- does anyone know why this error is popping up because i didn't have that error yesterday

[@melonchu3988]

everything was going smoothly until it said there was an error saying meta data generation failed...I'm not sure what that means or what do...

[@MichiganPeatMoss]

Mar 2023: This might wipe out the "industry" that sentence-mixing KZbin Poopers have accomplished for KZbin Poop (YTP) videos. :0

[@Squadala9001]

I've been making YTPs for over 15 years and have built up a reputation as one of the better sentence mixers out there and you're absolutely right - But I honestly don't see it as a negative, this technology is something I've dreamed of since I was a kid and I plan on fully embracing it. It's the next revolution in audio editing and you gotta adapt or die

[@illuminatiwave8093]

Nothing stops you from cloning voices, generating random phrases and cutting them like all life. You almost have the same process of cutting voices, only now you generate random things (I have always considered those cuts as the soul of YTPs. Removing them to put normal phrases takes away their essence.)

[@_Sepherial]

How do I use a cloned voice to read aloud a pdf file?

[@imanejoestar1578]

use a text to speech first

[@MeddlTitan]

Thanks so much easy explanation and easy to follow step by step you're doing gods work m8

[@Krikario]

Thank you so much. Been trying to find something like this for a while.

[@Tugezu]

What's the name of the first background song? Really groovy stuff! Also great tutorial :)

[@unamelable256]

For some reason - even through this video guide. Tortoise doesn't work because its screaming about torch module. Even when its installed already.

[@ihsacay]

i miss the tortoise-tts folder where do i find or download ?

[@FSXNOOB]

All seems to work, my own voice seems to be generated, but where do i listen to it?

[@Anonymous-ic3db]

No idea why I can't install this. Followed the setup instructions exactly as it was written and had nothing but errors.

[@sten40too1]

Does it work with Intel graphincs card?

[@andrewbanuelos7473]

what camera did you use to record your video? Looks super smooth.

[@bernayldrm2565]

I cant install pysoundfile. it says "conda is not recognized as an internal of external command, operable program or batch file." how can i solve this problem?

[@VAces]

i receive the following error code "PytorchStreamReader failed reading zip archive: failed finding central directory" can someon help please

[@reinermentges7816]

I re-started the anaconda prompt and i keep getting the error ModuleNotFoundError: No module named 'progressbar'

[@Nutronic]

How do u find Cuda version?

[@XooL1988]

Hi, it would be awesome to see a workflow, where the stuff is running in a docker :) I don't want to install all this on my real machine.

[@Adebup]

So we can't reproduce other accent? Like if I want someone to speak french, I just simply can't?

[@cosmicemotions-rk3nu]

Hey, quick question, is there a list of emotions you can run where you put [I'm exited]? Like is there only a select few that I can use? Thanks in advance 😀

[@Emit.Reviews]

You can put it anything you want. According to the developer himself, he didnt even know emotions could be replicated! Once people found out, he made so that putting words into brackets exclude them from the promt, but the AI still understands them. So this is a zone to explore :)

[@cosmicemotions-rk3nu]

@@Emit.Reviews Perfect, thanks!

[@Thisisjapan434]

for me it takes hours to export it. What can I do?

[@edharris9467]

Hello Emit. Thanks for this awesome tutorial. How can I find a list of different "tones" for the voices? Ex. Like you used [I'm excited] ? Thanks!

[@zicozichetto]

is possible generate a audio voice from a text with your own voice in mode free for language Italian?

[@kisho2679]

which website generates vocals: simulating female and male singer by just typing lyrics text, and specifying melody as well as pitch and key, and bpm?

0:47 you are not making a shrek is love reference in 2023

[@MindAtItsFinest]

Thanks bro now I can make Helen Mann voices for my MvM map

[@dhruvsaini9765]

Plz help sir!!! It doesn't clone hindi voice properly!!! 🙏

[@Emit.Reviews]

Yup as I mentioned in the video the model is trained on American voices. So it only works for English. You would need to find models that are trained in different languages (not sure for how to get it for Hindi)

[@ozzzyyyyy]

can we build for any language?

[@LyllianaofMirrah]

how tf can i get the 3.9 python version it's now using 3.10 which not compatible with the TorToise.

[@Emit.Reviews]

Download it manually and add it to path

[@LyllianaofMirrah]

@@Emit.Reviews hmm okay.

[@Emit.Reviews]

I have an updated version of this guide linked in the pinned comment. I showcase how to install everything after the updates

[@LyllianaofMirrah]

@@Emit.Reviews okay.

[@LyllianaofMirrah]

@@Emit.Reviews ahh yeah got it

[@23Tori0]

How do I get it to read an article? It seems to only be able to do less than 400 characters :(

[@WszystkoWiem]

I have RTX 3050 and ryzen 5 and 16 gb of ram but it loads very slowly. I wait about 20 minutes for a 5-word text.

[@matejsteinhauser3974]

I cannot download it. Cause it is now different. What do you say is no longer available. It is changed

[@jenzsharing8160]

Thanks for everything - so what is your recommendation for a voice cloning tool that keeps the accent. I have plenty of samples to feed it, but I would need it to accurately match the style of the original voice. Thanks for your help!

[@GinoDiCaprio]

What is the best way to clone my voice into an artificial intelligence voice synthesized by computers?

[@wildfuture.network]

Does it work for Mac too?

[@taraskhudiakov4825]

python: can't open file 'C:\\Users\\user\\tortoise\\do_tts.py': [Errno 2] No such file or directory

[@infinite5019]

thats what I'm getting

[@OwnyOne]

Anaconda prompt and navigator are not installing and this seems to be a very common problem for which I haven't yet found the solution to :')

[@patrick5044]

I found the prompt on the start menu by refreshing it. All I did was closing and reopening explorer.exe and you could achieve the same by rebooting the system.

[@erm2215]

Dear sirs..how to fix "no such file / direction"

[@madara_u_chiha]

FFS tts doesn't work on my mac

[@jimitnahar1714]

can we do this if we don't have NVIDIA graphic card?

[@infinite5019]

it keeps saying can't find main module

[@НастяСироткина-з5с]

Hello, could you say how many languages does it support?

[@erikdong]

Brrrrriiiiillllant. great info and process breakdown. Thank you for sharing!

[@BrennerEraFan]

Can this also be done on mac?

[@latrechedhiyaeddine]

thanks for video that really cool but what about other languge can i use them normall like arabic and franch and thanks you really cool

[@bestvideogamer4560]

im getting the error: "UserWarning: No audio backend is available."

[@lanceskrill4599]

Can i use java at 01:27 , im tyred of python not opening and crashing and 38 other errors , like not having torch installed , the 7z archive and 7z couldnt extract some essential dll file .

[@009badboy]

is there any tool that you can clone voice without english accent in ither language than english and them make it work not from text but from talk to microphone?

[@tharana5167]

Hey quick question, can we use this for KZbin video for free

[@dr.killjoy667]

I'm getting this error "File "C:\Users\Dons\anaconda3\lib\site-packages\pydantic-2.0a4-py3.10.egg\pydantic\__init__.py", line 1, in from pydantic_core import ValidationError ModuleNotFoundError: No module named 'pydantic_core'" anyone know why I'm getting this?

[@kwiatu7188]

same do you know how to fix?

[@dr.killjoy667]

@@kwiatu7188 I gave up for now.

[@Fynmarr]

@@dr.killjoy667 Same problem. Be nice if someone knew what was up.

[@Fynmarr]

Got it. type in "conda install pydantic". All done.

[@dr.killjoy667]

I got it working, thank you

[@inachu]

Trying to learn how to do this so I can make a voice but use the voice from that serious sounding dude from frontline

[@baxe3186]

is this working with like movie characters i want to voice them

[@shinkkk]

Alright got errored lamo

[@forgot2799]

any help would be appreciated, installed it like the video OMP: Error #15: Initializing libiomp5md.dll, but found libiomp5md.dll already initialized. OMP: Hint: This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results.

[@lindend7984]

same

[@forgot2799]

@@lindend7984 figured it out just go to environment variables then go down into system variables and press new and add 'KMP_DUPLICATE_LIB_OK=TRUE(the value is true) and you should be good

[@lindend7984]

@@forgot2799 thanks for your help!

[@mcchunkz9792]

@@forgot2799 how do u do that?

[@soundmelody-vlogbackground579]

The video is good. But I think you need to zoom it, a litle bit more... thanks.

[@phantomBlurrrrr]

Thank you, it works

[@drormo]

is there a docker version of this ?

[@intell0]

Will this work on Mac?

[@FahadKhan-vx6oj]

how do I change my voice

[@jamesjames8805]

it sounds really realistic . I think it's same with 11Labs voice

[@Reckless_Aura]

Is there anyone that can help with this please? Could not open requirements file: [Errno 2] No such file or directory: './requirements.txt'

[@Emit.Reviews]

Are you pasting in all 4 lines at the same time?

[@МойВолк-ч8ъ]

Basically that means that the file "requirements.txt" does not exist in the location you are searching for it. You likely are not in the correct directory when you execute the command. If you execute a script that calls that command, make sure that you execute the script in the correct location.

[@Reckless_Aura]

@@МойВолк-ч8ъ I fixed the problem there was an extra line of code that I missed that I had to execute but I am getting another error

[@OLPTreeMan-tv4uq]

Can I do all of this offline

[@TuyenTran-zy6ze]

dear sir, this way can make for other language, not only english, right? i try to install and make voice other my language for vietnamese but can not, pls help me! thanks!

[@Emit.Reviews]

Im pretty sure this model was trained on English voices. So other languages arent supported

[@alexkelly757]

does anyone know of any models for British accents?

[@Kawayoporu]

I hope you can clone a voice from another language to English right? Because I was thinking of trying to dub a cartoon clip in English.

[@saulboghiu9035]

Would this still work if I have CUDA version 12.1?

[@matejsteinhauser3974]

Can you please give us toolbox already. Dont be indifferent to me because I cannot do coding and it always changes over time

[@SnackCityMafia]

Can you run this on colab?

[@ItzA2DaWAnimate]

mine says failed what do i do?

[@CrimeanHorseArcher]

Is there a way to imitate a persons voice in another language? (Dutch) One of our female actors died and we need to cover her voice for a couple of sentences in a badly recorded scene

[@Emit.Reviews]

Sorry to hear that! While this model probably wont work. I highly recommend trying out eleven labs (its around $10/mo). You can upload voice samples and play around with some settings. They have multilanguage support (Dutch not added in yet) but you might get some decent results either way

[@kalebs_shaft]

can i use miniconda?

[@mastersteamer3697]

it doesnt let me pick notepad++, the next button is just grayed out

[@Emit.Reviews]

Doesnt really matter. It might have been because i had notepad++ installed. This setting doesnt rly come into play that much

[@JM-lw1dr]

anyone get the no module named progress bar error

[@Emit.Reviews]

Check out the pinned comment. I show an updated method, this one seems to glitch out after Python update

[@UBAID31]

Can we do it on Android?

[@amiiinafedora2318]

I'm getting a ModuleNotFoundError: No module named 'torchaudio' Any way to fix this?

[@Emit.Reviews]

Check the pinned comment

[@dr.imayavarambanmunuswamy808]

Many thanks Sir. Great.

[@botlifegamer7026]

OMG I finally got it working had to delete all previous installs lol and had to use Numpy 1.24.3

[@hamza_clipz_699]

Do I have to have nvdia graphics card for this

[@Emit.Reviews]

Yes, it uses cuda core tech which only nvidia cards have

[@hamza_clipz_699]

@@Emit.Reviews so tortoise cant be used without nvdia, correct?

[@Emit.Reviews]

Yes, at least locally. I haven't tried using it non-locally but it might be possible.

[@nonameman7114]

@@Emit.Reviews any alternatives for people that don’t have Nvidia ?

[@lunarusdust7227]

@@Emit.Reviews is there a way you can update this video so we non nvidia users can have a shot?

[@TheDukeOfZill]

I tried to do the apple tree from the 1939 film the wizard of oz.. wasn't what i hoped for :( darn lol. i really really want to find a way to use that voice exactly as it is -_-

[@xanderevert-so6ln]

it this safe?

[@Altinferno]

Barry is love. Barry is life.

[@fighteraircraft4576]

Can it be done on AMD GPU?

[@SneedFeedSeed]

no

[@hold-ulv3552]

hi Emit can you make a video about x10 hosting and ultifreehosting as i won't to see if is good to use

[@Emit.Reviews]

I've tested them out before and didn't really like them. I've also made a video on free web hosting that you can find here (the best option was the oracle free tier) kzbin.info/www/bejne/qKOro5aIgreHqrM&

[@ffiloi]

does this work in spanish? or is it only for english voices?

[@asmundhessevikeikeland7346]

can you use norwegian voice

[@stevenw5253]

can't wait to troll my friends

[@ashtrixlabs]

This is working, thank you for the tutorial. But it's taking 10 minutes to generate a simple text input: "I'm going to speak this" on my machine. Is it my hardware or can I do something to decrease this output time. Again, thanks for the video, it was really a fun process

[@Emit.Reviews]

Thats why its called "tortoise" haha. Its a slow model by design. There is a new version called tortoise 5x faster. Ill create a tutorial on that later

[@wildfuture.network]

@@Emit.Reviews is the faster one also Free?

[@wildfuture.network]

Actually good for me that it'd make me sound American since I wanted it to be pretty much my voice but without the annoying european accent so, i do like that!!

[@forsert7679]

was that a shrek is love referance?

[@phen-themoogle7651]

I'm trying to figure out how to make a voice but for Japanese language....so hard :( All I could do in python was google's TTS Japanese voice. I wanna make something unique but realize it'll take tons of machine learning or knowledge of voice synthesis or stuff above my level...*sigh* I don't wanna pay monthly fees for stuff like GPT4 or ElevenLabs (and Eleven Labs didn't even have Japanese voices the last time I checked)...