One click pipeline would be awesome

Jarod I just found this channel and you are going to blow up man you earned it. Thank you for existing.

Excellent video man, can't wait for the 1-click pipeline that would be awesome !

This is so funny I saw every ajatt channel in the 2-3 years ago except yours.

This needs to be updated now. TorToiSE-TTS-Fast has gained the 5x GPU performance lift in training that 11Labs uses. 11Labs is still REALLY fast for training. Did they implement some of that "5-second voice clone source sample" training technology that was talked about online before? Also, what's the deal with the current model generation creating such noisy, muted output? UVR is okay, but it would probably be of benefit to add an EQ filter pass to the output to enhance it. And what about higher sample rates for output? Can we resample it to CD-quality 44.1KHz?

This is really fantastic.

your a legend mate

Hey jarod, thats amazing... crazy to see how good the ai voice can sound like with a proper technique. How is the process goingt to merge the worksteps? keep up the great video quality. you´re quite helpfull! :)

That's crazy good thank you.

Thanks man everything works.

super good info!

This is really fantastic. The one thing that would make this whole system truly out of this world would be to be able to use multiple GPUs on each step. Good sir Jarod, do you know if it’s possible or how to generate on tortoise with multiple GPUs? I’m searching for this answer but I’ve yet to find it😅

Thanks for this idea Jarod! i have a issue though. when i generate clips from tortoise tts mrq, then use those clips inside RVC, RVC changes the voice a tiny bit.. how can i make sure RVC doesn't change the way words are said? I am using the same voice from tortoise and in RVC.

Great content, thanks. Could you please post a link to a video about the audio book creator you reffered to in the beginning of your video. I could'nt find it on your channel.

what does the new setting in the ai voice changer mean? its under input and output and its called monitor

you can use tortoise trained models interchangeably with RVC?

I've finally gotten everything working but have noticed when it slices the audio for training it cuts it into 57 small files so it only trains from a small portion of the full audio. Also no matter how long I train for, when I finally do a text-to-speech generation it never really sounds anything like the device and either has an American accent or a really posh sounding southern-UK one. (I'm from northern england).

Hey! Are there alternatives to RVC? Which is the best for TTS?

I set up the program the right way and it worked fine for awhile but then I got the lastest update to 3.8 version and the audio won't play threw any of the applications but I can still here it otherwise?

How did you clone your voice?! I was training with 8 hours voice sample for 12 hours and got horrible resaults…

i've been playing around with RVC and this is exactly the sort of thing i was thinking of trying. well, first i wanted to see what it would do with audio from my melodica, THEN this. anyway, i don't want to keep paying elevenlabs with their unfair character resets after the end of the month and total lack of control. ideally i would want to try this with bark instead, since bark gives way better control than tortoise. but i'm just too mentally drained from work and all sorts of projects i have going (which includes learning to read japanese which so far has been more fun and way easier than i thought it would be, making a visual novel -> which is why i need the AI voices, and fixing my broken 3d printer) to try to get bark and tortoise to work. i know it won't be difficult once i try, but the activation energy is more than i have right now.

is there a way to do the reverse? using RVC models as voices for tortoise?

I will give Tortoise a chance, thank you

Definitely a pipleline would be good - click no - would like to be able to run this from other software for complete automation. Would also be nice to be able to handle segmenting so you could give it a long text input file and come out with the spoken version. Thanks.

wanted to ask if you could maybe upload the voice data of some vtubers so i can have more in my voicechanger

i already have an rvc model of my friend and now i want to use tortoise

Hello help me error (When start Training get an Error: (result, consumed) = self._buffer_decode(data, self.errors, final) UnicodeDecodeError: 'utf-8' codec can't decode byte 0x84 in position 0: invalid start byte) If you use the learning rate scheduler (calling scheduler.step()) before the optimizer’s update (calling optimizer.step()), this will skip the first value of the learning rate schedule. If you are unable to reproduce results after upgrading to PyTorch 1.1.0, please check if you are calling scheduler.step() at the wrong time.

Hi Jarod. Thank you for this. I have one question in TTS when I generate the voice and press play to hear it inside of TTS it sounds good but then when I download it and try to play it, it sounds much worse with a lot of static. Why is this happening?

Hey there, Jarod! Back again stalking your comment section👋😊 I was wondering whether this solution has any good shot by running it in the cloud? Thanks! P.S. your channel is going to blow up very soon 💪💯

Thanks for the amazing vids! What are your thought on YourTTS? and compared to Tortoise TTS?

could you teach ups how to install Tortoise TTS + RVC so it looks like yours

If i trained a model in tortoise, can i use it in rvc?

Can i do this for any language?

Great

Is there a website similar to civitAI but for voice cloning models?

How do I use the .pht files that I have downloaded for Tortoise TTS + RVC? Where should I place them so that the voices can be loaded in? I found the solution: The .pht needs to be in a map with at least audio file before you can run (Re)Compute Voice Latents and generate audio. File Location = C:\YourPathLocation\ai-voice-cloning\voices

Hello, really thanks for your guides! Can you please say what is the name of a microphone you use?

I kinda hoped there's also a version for google colab, bc i cant really run it 😅

is there a community for tortoise tts? i need to ask question

Hello Mr Jarods Can you please tell me How to make dubbed video for other language with voice cloning and TTS with simulation feelings and emotions for example 3 mins of anime or movie ❤❤❤ Thank you

I once listened to Ina (from Hololive) reading the Spice and Wolf Light novel, she didn't do the entire series and I really wish she did. With this kind of thing, that will be possible- albeit with a lack of emotion.

How to do I get the voice changer todo work on discord

I have problems connecting my sony Wh-100XM4

Only work with voices in english?

Can Tortoise tts be used for commercial purposes i.e. in KZfaq videos, Tiktok etc?

Hey! I appreaciate all your videos. I was wondering if you would make a tutorial on how to make any AI model to sing songs?

Did you finish the one click pipline?

if theres a comfy ui style pipeline for this that would be cool

Can I use it on Windows 11?

Can you also teach us how to remove it?

Is i5 12500h with rtx 3050 enough? Im new to a.i and really captivated learning a.i

Okay.. I do watch so many of your videos, I have 1 question though.. I really like to run it on my local computer instead of google colab if possible. Is RTX 2070 Super is enough for running all of this? I don't mind waiting 24 hours of training etc. Is it possible? Or it is not enough? Thank you for sharing these..

OK, this was hilarious because even after you switched back to your real voice, I thought it was the AI voice still LOL. I then tried it in reverse, and the AI voice sounded like your real voice to me. (funny how I get tricked just by the order)

Awesome tutorial! Thanks for spending $5 lol

I’ve fallen down a rabbit hole, catching up but it’s a lot to take in, i’m going to try all you’ve taught, even then though… once you’ve come up with your “one click pipeline”, assuming it’ll be as if you’re selling an easy button to all this, you have a buyer here, and i’m sure you’ll make a sh** ton

There is any multilanguage opensource model?

Is there any way to load in pre trained models for tortoise? I haven't been getting very favorable training results that I'm working on improving, hut am curious if this is possible in the meantime

can you make a colab tutorial aswell?

How good does it perform with multilingual trasnformations?

does anyone know any good colabs for tortoise tts?

Is it possible yo use RVC models on Tortoise for TTS? If not, any good ppaces to download TTS models?

You should try 11 labs with the v2 model. The similarity is a LOT better. I can actually clone anime English dub voices which I couldn’t do before with v1. Tortoise never really worked for me but you seem to have figured it out. RVC is great. I have made some funny covers with it

So I have yet to get RVC to accept any of my trained voices from Tortoise TTS. They work in Tortoise just fine, but when I try to load them into RVC I get attrobuteerror's. I have tried both RVC1 and 2. :/ Also.. Love your videos! They have been a big help in my AI journey!

I'll try to connect Tortoise to RVC.

It has a major issue, which is the inability to fine-tune languages other than English. I've tried to address this problem by replacing the pre-trained model and tokenizer, but it resulted in various errors. Finally, I came across the author's statement that the model is trained specifically for English, and if we want to use models for other languages, we would need to retrain them within this software, which presents significant challenges in terms of data quantity and time requirements.

Please help me i get an error that says torchaudio not found

I'm a little confused, is RVC a TTS thing?

WTF, Marine Senpai speaks English !

where's the link?

...if it can't properly say AI or other words, just phonetisize it with word's/spellings that would create the word phonetically. "Ay Eye."

Can we make Hindi language?

first!

17 Seconds after upload :P But still too slow for first comment @@Afflictionability

your videos are long and say nothing. same content in 3 mins other videos

Awesome video, thank you. I'm wondering is there a website were you can download trained voice models already? I only found one on huggingface.

RVC Please answer my question, do you have Telegram or Facebook to communicate with you? I trained an audio clip and reduced it to 10 seconds and trained it. Do you know how long it takes for my laptop Lenovo core i5 10th generation RAM 12 and graphics card amd Does it succeed in training Please answer my question

Please answer my question, do you have Telegram or Facebook to communicate with you? I trained an audio clip and reduced it to 10 seconds and trained it. Do you know how long it takes for my laptop Lenovo core i5 10th generation RAM 12 and graphics card amd Does it succeed in training Please answer my question