rsxdalv / tts-generation-webui Goto Github PK

Hi guys, I love the project, this is not an issue just a question, How can I query it with http? There is any api flag or something? Thanks for taking the time :3

This script relies on Miniconda which can not be silently installed under a path with spaces.
Press any key to continue . . .

I am coming across the following error when Starting. I can see the WebUI but get errors when trying to run cloning or anything else. When Cloning I get this error: RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory |

And when running Totoise I am getting the following:

2023-07-30 21:14:38 | WARNING | xformers | WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:
PyTorch 2.0.1+cu118 with CUDA 1108 (you have 2.0.1+cpu)
Python 3.10.11 (you have 3.10.12)
Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers)
Memory-efficient attention, SwiGLU, sparse and more won't be available.
Set XFORMERS_MORE_DETAILS=1 for more details
2023-07-30 21:14:39 | WARNING | xformers | Triton is not available, some optimizations will not be enabled.
This is just a warning: No module named 'triton'

Any solutions? I am not a coder, so go gentle. Thanks in advance

I used the latest installation package “one-click-installers-tts-4.0” for installation, but encountered an error during the final installation.

Is this error related to an exception during installation? How can I solve this problem?

Have you considered adding Tortoise-TTS-Fast instead of the original Tortoise-TTS? Fast performs much faster, and you can add your own models to it. It's better to clone voices because you can extract the latents from whole audio samples, making them more sophisticated.
https://github.com/152334H/tortoise-tts-fast

would it be posibel to add the german huber model version from C0untFloyd (Model) to the Bark Voice Clone Tab.

Some HuggingFace accounts appear to have issues, as a result if you see errors relating to 404 or 401 when downloading the MusicGen model, this might be useful.

pip uninstall audiocraft
pip install git+https://[email protected]/GrandaddyShmax/audiocraft_plus@85a5112#egg=audiocraft

Or just edit ..\env\Lib\site-packages\audiocraft\models\loaders.py to use GrandaddyShmax repo

HF_MODEL_CHECKPOINTS_MAP = {
    "small": "GrandaddyShmax/musicgen-small",
    "medium": "GrandaddyShmax/musicgen-medium",
    "large": "GrandaddyShmax/musicgen-large",
    "melody": "GrandaddyShmax/musicgen-melody",
}

Originally posted by @rsxdalv in facebookresearch/audiocraft#139 (comment)

I tried use RVC Beta Demo so i get this error.
"RuntimeError: "slow_conv2d_cpu" not implemented for 'Half'"
How can i fix this?

Specifications:

• f0 Up key = 0
• f0 Method = harvest
• Index Rate = 0,66
• Filter Radius = 3
• Resample SR = 0
• RMS Mix Rate = 1
• Protect = 0,33

Hubert
-Device = CPU (I did the installation for "CPU only")

• Use half precision model (Depends on GPU support) = [ ](not check)

Is it possible to choose where cached stuff goes to?
My OS Install or C: is becoming full and I ordered a dedicated external HDD for AI stuff.
For example, the models go to my user cached directory, I want everything to go to my separate HDD in a certain folder.

when i press on the icon to upload i receive this error
File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1352, in process_api
result = await self.call_function(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1077, in call_function
prediction = await anyio.to_thread.run_sync(
File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
result = context.run(func, *args)
File "/content/tts-generation-webui/src/tortoise/gr_reload_button.py", line 24, in
fn=lambda: open_folder(dirname),
File "/content/tts-generation-webui/src/history_tab/open_folder.py", line 12, in open_folder
subprocess.check_call(["xdg-open", "--", folder_path])
File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['xdg-open', '--', '/content/tts-generation-webui/data/models/rvc/checkpoints']' returned non-zero exit status 1.
xdg-open: unexpected option '--'
Try 'xdg-open --help' for more information.
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1352, in process_api
result = await self.call_function(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1077, in call_function
prediction = await anyio.to_thread.run_sync(
File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
result = context.run(func, *args)
File "/content/tts-generation-webui/src/tortoise/gr_reload_button.py", line 24, in
fn=lambda: open_folder(dirname),
File "/content/tts-generation-webui/src/history_tab/open_folder.py", line 12, in open_folder
subprocess.check_call(["xdg-open", "--", folder_path])
File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['xdg-open', '--', '/content/tts-generation-webui/data/models/rvc/checkpoints']' returned non-zero exit status 1.

where is the voice clone tab? I cab see it on the git page, but its not there on my webui?

one click installer fails for M2

he following NEW packages will be INSTALLED:

  ffmpeg             pkgs/main/osx-arm64::ffmpeg-4.2.2-h04105a8_0
  gnutls             pkgs/main/osx-arm64::gnutls-3.6.15-h887c41c_0
  lame               pkgs/main/osx-arm64::lame-3.100-h1a28f6b_0
  libidn2            pkgs/main/osx-arm64::libidn2-2.3.4-h80987f9_0
  libopus            pkgs/main/osx-arm64::libopus-1.3-h1a28f6b_1
  libtasn1           pkgs/main/osx-arm64::libtasn1-4.19.0-h80987f9_0
  libunistring       pkgs/main/osx-arm64::libunistring-0.9.10-h1a28f6b_0
  libvpx             pkgs/main/osx-arm64::libvpx-1.10.0-hc377ac9_0
  nettle             pkgs/main/osx-arm64::nettle-3.7.3-h84b5d62_1
  openh264           pkgs/main/osx-arm64::openh264-1.8.0-h98b2900_0
  x264               pkgs/main/osx-arm64::x264-1!152.20180806-h1a28f6b_0

The following packages will be DOWNGRADED:

  cryptography                       41.0.2-py310h6204c90_0 --> 41.0.2-py310h6e31b35_0
  curl                                     8.1.1-h80987f9_1 --> 8.1.1-h80987f9_0
  git                               2.40.1-pl5340h6cf2078_1 --> 2.40.1-pl5340h3afa44c_1
  krb5                                    1.20.1-hf3e1bf2_1 --> 1.19.4-h8380606_0
  libcurl                                  8.1.1-h3e2b118_1 --> 8.1.1-h0f1d93c_0
  libnghttp2                              1.52.0-h62f6fdd_1 --> 1.52.0-h10c0552_1
  libssh2                                 1.10.0-h02f6b3c_2 --> 1.10.0-h449679c_2
  openssl                                  3.0.9-h1a28f6b_0 --> 1.1.1u-h1a28f6b_0
  python                                 3.10.12-hb885b13_0 --> 3.10.12-hc0d8a6c_0

Preparing transaction: done
Verifying transaction: done
Executing transaction: done
Traceback (most recent call last):
  File "/Users/kibotu/Documents/repos/ai/tts-generation-webui/webui.py", line 96, in <module>
    install_dependencies()
  File "/Users/kibotu/Documents/repos/ai/tts-generation-webui/webui.py", line 60, in install_dependencies
    update_dependencies()
  File "/Users/kibotu/Documents/repos/ai/tts-generation-webui/webui.py", line 68, in update_dependencies
    os.chdir("tts-generation-webui")
FileNotFoundError: [Errno 2] No such file or directory: 'tts-generation-webui'

The installer works fine for windows but I get this error when installing on macOS:

Retrieving notices: ...working... done
ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'
Traceback (most recent call last):
  File "/Users/Downloads/one-click-installers-tts-4.0/webui.py", line 125, in <module>
    install_dependencies()
  File "/Users/Downloads/one-click-installers-tts-4.0/webui.py", line 63, in install_dependencies
    update_dependencies()
  File "/Users/Downloads/one-click-installers-tts-4.0/webui.py", line 77, in update_dependencies
    os.chdir("tts-generation-webui/models/bark")
FileNotFoundError: [Errno 2] No such file or directory: 'tts-generation-webui/models/bark'

After manually adding the required directories and files, I get this error:

/System/Volumes/Preboot/Cryptexes/OS/Users/Downloads/one-click-installers-tts-4.0/installer_files/env/lib/python3.10/site-packages/torch/lib/libtorch_global_deps.dylib' (no such file), 
'/Users/Downloads/one-click-installers-tts-4.0/installer_files/env/lib/python3.10/site-packages/torch/lib/libtorch_global_deps.dylib' 
(mach-o file, but is an incompatible architecture (have 'x86_64', need 'arm64'))

Hi again.
I'm using your TTS 4.0 on a daily basis.
I think its the most comprehensive and usable so far.
I ran the update_windows.bat to get the latest.
I saw that it brought in some changes, however when I later tried to start the app I ran into the following:

+++++++++++++++++++++++++++++++++++++
from src.tortoise.gen_tortoise import generate_tortoise_long
File "C:\SuperStableDiffusion2.0\TTS-4.0\tts-generation-webui\src\tortoise\gen_tortoise.py", line 7, in
from tortoise.api import TextToSpeech, MODELS_DIR
File "C:\SuperStableDiffusion2.0\TTS-4.0\installer_files\env\lib\site-packages\tortoise-2.4.2-py3.10.egg\tortoise\api.py", line 14, in
from tortoise.models.autoregressive import UnifiedVoice
File "C:\SuperStableDiffusion2.0\TTS-4.0\installer_files\env\lib\site-packages\tortoise-2.4.2-py3.10.egg\tortoise\models\autoregressive.py", line 6, in
import deepspeed
ModuleNotFoundError: No module named 'deepspeed'

Done!
Press any key to continue . . .
+++++++++++++++++++++++++++++++++++++++

I opened cmd and tried running >pip install deepspeed
but it cried out for torch, (which I know is already installed) but didn't want to stuff it up - so I decided to leave it.
Your thoughts?

I am having an issue setting a username and password to the Gradio interface on this program. In the webui it allows the user to specify the username and password associated with the Gradio UI. I set the auth as username:password as Gradio Docs suggest to no avail. The error I get is when I restart and go to log into the webui. It will always display the incorrect login error. I tried to configure the same settinng in the config.json file and --gradio-auth-path flag with no changes. Is this something that I am doing wrong or is there a bug in the authentication page of this project?

-theman23290

Specs:
OS: Debian 11
CPU: Xeon E5 2670 v3 (25C)
RAM: 80GB
GPU: None

When I launch I get this message:

No GPU being used. Careful, inference might be very slow!

I also get this message earlier in the log, though I am not sure it is related:

WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:
    PyTorch 2.0.1+cu118 with CUDA 1108 (you have 2.0.1+cpu)
    Python  3.10.11 (you have 3.10.11)
  Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers)
  Memory-efficient attention, SwiGLU, sparse and more won't be available.
  Set XFORMERS_MORE_DETAILS=1 for more details

I indicated option "A" when I installed but it didn't seem to get my 4090 GPU. Should I try reinstalling, or is there any quicker way to try to repair the GPU detection issue?

Hello there! Firstly, I must say that this web UI is absolutely amazing! Great job!
I would like to inquire about how I can choose a specific voice for the text-to-speech (TTS) generated by Bart. Currently, it seems to use random voices.
Thank you for your assistance.

Using the one-click installer, I got errors launching the webUI after installation:

Traceback (most recent call last):
  File "/home/***/tts-generation-webui/tts-generation-webui/server.py", line 40, in <module>
    from src.tortoise.generation_tab_tortoise import generation_tab_tortoise
  File "/home/***/tts-generation-webui/tts-generation-webui/src/tortoise/generation_tab_tortoise.py", line 2, in <module>
    from src.history_tab.open_folder import open_folder
ImportError: cannot import name 'open_folder' from 'src.history_tab.open_folder' (/home/***/tts-generation-webui/tts-generation-webui/src/history_tab/open_folder.py)

I resolved this by changing elif sys.platform == "linux2": to elif sys.platform == "linux":" in src/history_tab/open_folder.py. Maybe something to do with how python recognises WSL2? Anyway that fixed it.

Trying this on an Nvidia GPU, the 1650 Super to be exact. The entire installation process seemed to have gone fine, I selected Nvidia when asked. No issues.

However, at the end I got this error:

Env file not found. Creating default env.
Config file not found. Creating default config.
Traceback (most recent call last):
  File "C:\Users\USER\Desktop\one-click-installers-tts-6.0\one-click-installers-tts-6.0\tts-generation-webui\server.py", line 40, in <module>
    from src.tortoise.generation_tab_tortoise import generation_tab_tortoise
  File "C:\Users\USER\Desktop\one-click-installers-tts-6.0\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\generation_tab_tortoise.py", line 7, in <module>
    from src.tortoise.gen_tortoise import (
  File "C:\Users\USER\Desktop\one-click-installers-tts-6.0\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\gen_tortoise.py", line 7, in <module>
    from tortoise.api import TextToSpeech, MODELS_DIR
ModuleNotFoundError: No module named 'tortoise'

Done!

Then it said press any key to continue. The instructions.txt isn't really clear on how to run the actual webui. So I tried running the webui.py and it just says Conda is not installed. Exiting... in the terminal and doesn't run.

I then tried to run the server.py just in case, and for some reason it said I didn't have gradio installed. So I went ahead and installed gradio just by doing pip install gradio. This let the script run, but similar to the one click installer terminal error, it can't find tortoise. I'm also not sure if that's the only issue, or if the one-click installer wanted to continue doing things after that, but stopped when that failed.

A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
Loading extensions:
Loaded extension: callback_save_generation_ffmpeg
Loaded extension: callback_save_generation_musicgen_ffmpeg
Loaded extension: empty_extension
Loaded 2 callback_save_generation extensions.
Loaded 1 callback_save_generation_musicgen extensions.
Loading Bark models
- Text Generation: GPU: Yes, Small Model: Yes
- Coarse-to-Fine Inference: GPU: Yes, Small Model: Yes
- Fine-tuning: GPU: Yes, Small Model: No
- Codec: GPU: Yes
Traceback (most recent call last):
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\server.py", line 13, in
from src.bark.generation_tab_bark import generation_tab_bark
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\src\bark\generation_tab_bark.py", line 27, in
from src.model_manager import model_manager
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\src\model_manager.py", line 4, in
model_manager = BarkModelManager(config)
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\src\bark\BarkModelManager.py", line 8, in init
self.reload_models(config)
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\src\bark\BarkModelManager.py", line 29, in reload_models
preload_models(
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\models\bark\bark\generation.py", line 327, in preload_models
_ = load_model(
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\models\bark\bark\generation.py", line 275, in load_model
model = _load_model_f(ckpt_path, device)
File "D:\AI\TTS\one-click-installers-tts-4.0\tts-generation-webui\models\bark\bark\generation.py", line 240, in _load_model
model.to(device)
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 1145, in to
return self._apply(convert)
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 797, in _apply
module._apply(fn)
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 797, in _apply
module._apply(fn)
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 797, in _apply
module._apply(fn)
[Previous line repeated 2 more times]
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 820, in _apply
param_applied = fn(param)
File "D:\AI\TTS\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 1143, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 6.00 GiB total capacity; 5.25 GiB already allocated; 0 bytes free; 5.35 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

Hi! I have a error when i upload sample of my voice. What I'm doing wrong?

EDIT: Nevermind. I was commented this line in files.

I was making my first test with tortoise model and this started to download all requirements to generate the audio but my hard disk get full and this operation could not be finished and get me an error, so I restarted the application and the error still were there, so I delete all the folder for trying to fix the problem, then I reinstall everything once more but the issue still keeps, that means that the installation of the regresive model of tortoise is in a folder out of the application, can you say me where can I find this corrupted file?

Generating tortoise with params:
TortoiseParameters(
text='hello world',
voice='random',
preset='ultra_fast',
seed=-1.0,
cvvp_amount=0.0,
split_prompt=False,
num_autoregressive_samples=16,
diffusion_iterations=30,
temperature=0.8,
length_penalty=1.0,
repetition_penalty=2.0,
top_p=0.8,
max_mel_tokens=500,
cond_free=True,
cond_free_k=2,
diffusion_temperature=1.0,
model='Default'
)
Traceback (most recent call last):
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\blocks.py", line 1352, in process_api
result = await self.call_function(
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\blocks.py", line 1093, in call_function
prediction = await utils.async_iteration(iterator)
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\utils.py", line 341, in async_iteration
return await iterator.anext()
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\utils.py", line 334, in anext
return await anyio.to_thread.run_sync(
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 807, in run
result = context.run(func, *args)
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\gradio\utils.py", line 317, in run_sync_iterator_async
return next(iterator)
File "C:\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\generation_tab_tortoise.py", line 162, in gen
yield from generate_tortoise_long(
File "C:\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\gen_tortoise.py", line 188, in generate_tortoise_long
datas = generate_tortoise(
File "C:\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\gen_tortoise.py", line 100, in generate_tortoise
tts = get_tts()
File "C:\one-click-installers-tts-6.0\tts-generation-webui\src\tortoise\gen_tortoise.py", line 77, in get_tts
MODEL = TextToSpeech(
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\tortoise\api.py", line 237, in init
self.diffusion.load_state_dict(torch.load(get_model_path('diffusion_decoder.pth', models_dir)))
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\torch\serialization.py", line 797, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "C:\one-click-installers-tts-6.0\installer_files\env\lib\site-packages\torch\serialization.py", line 283, in init
super().init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

Hi, voice cloning is working good with bark, is there way of training more expressions in Bark?

Is this two separate things, "bark" and "tortoise" or are you planning to somehow combine them so that "bark" can use the "tortoise" model?

This is not an issue as such but a request for advice about where to set up an installation.

Firstly, your idea to create a cross-platform tts tool is extremely valuable, especially with a one-click installer. however
I am hesitant to just set up a folder in my C:\Users\NAME drive and expect everything to go flawlessly.

Over the past year the advent of Ai generated tti and LLM's has been an exciting journey.
I am one of those people who has gone through a steep learning curve getting to grips with virtual python environments and do not fully understand the intricacies of how the multitudes of modules inter-relate.
I do know enough to know that that conflicts can occur between them and that a lot of time can be spent un-installing and reinstalling them.

At present I have successfully set up several tts applications: Coqui, Silero, & Bark.
I have also attempted to set up Tortoise and AudioCraft, but have failed to troubleshoot installation errors. (Note1. below)

I first started using Silero in oobabooga, which worked fine.
However the voices were v. limited so I set up coqui and bark to get a better variety of voices and accents.
And the possibility of voice cloning/training is extremely appealing also.

All that is a long way to ask the question:

1. Is there an ideal place to install your tts-webui that will not create conflicts with other installations?
2. Should I un-install all the other applications first?
3. Is it likely that the installation will add duplicate versions of torch, conda and other dependencies that are already installed?

BACKGROUND
I am using Windows 10, RTX 4070Ti, CUDA 11.7, Anaconda3, conda 22.9.0, Torch 2.0.1
2023-06-25_pip list.txt

Here is a list of the folders where I have set up various applications:
Initially I installed tts so that they could be used with oobabooga and later Silly Tavern
C:\SuperStableDiffusion2.0\stable-diffusion-webui
C:\SuperStableDiffusion2.0\oobabooga-windows
C:\SuperStableDiffusion2.0\Bark\bark-gui
C:\SuperStableDiffusion2.0\CoquiTTS\TTS

Later I began setting up applications in the User\ directory
C:\Users\ABC\Audiocraft\audiocraft-main
C:\Users\ABC\Bark-tts\bark_win\bark-gui
C:\Users\ABC\Silero-tts
C:\Users\ABC\coqui-tts
C:\Users\ABC\tortoise-tts

(Note 1.) Issues Raised:
neonbjb/tortoise-tts#468
facebookresearch/audiocraft#123

You might or might not be aware of gitmylo/audio-webui which has very similar goals. Maybe it would make sense to combine forces with gitmylo?

https://github.com/gitmylo/audio-webui

After installing I get a few errors and then it crashes

A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
Loading extensions:
Loaded extension: callback_save_generation_ffmpeg
Loaded extension: callback_save_generation_musicgen_ffmpeg
Loaded extension: empty_extension
Loaded 2 callback_save_generation extensions.
Loaded 1 callback_save_generation_musicgen extensions.
Loading Bark models
- Text Generation: GPU: Yes, Small Model: Yes
- Coarse-to-Fine Inference: GPU: Yes, Small Model: Yes
- Fine-tuning: GPU: Yes, Small Model: No
- Codec: GPU: Yes
Traceback (most recent call last):
File "C:\Projects\AI\audiogens\one-click-installers-tts-4.0\tts-generation-webui\server.py", line 14, in
from src.bark.clone.tab_voice_clone_demo import tab_voice_clone_demo
File "C:\Projects\AI\audiogens\one-click-installers-tts-4.0\tts-generation-webui\src\bark\clone\tab_voice_clone_demo.py", line 7, in
from models.bark_voice_cloning_hubert_quantizer.hubert.pre_kmeans_hubert import (
File "C:\Projects\AI\audiogens\one-click-installers-tts-4.0\tts-generation-webui\models\bark_voice_cloning_hubert_quantizer\hubert\pre_kmeans_hubert.py", line 16, in
import fairseq
ModuleNotFoundError: No module named 'fairseq'

Done!
Press any key to continue . . .

This is the output on launching start_windows.bat

Using the v3.1 one click installer, I had to comment out lines 12 and 126 of tts-generation-webui/server.py to get the start_windows.bat to work. Ran a few tests and got great results with both Bark and Tortoise.
After closing and restarting the webui, i got the error: ValueError: not enough values to unpack (expected 2, got 1). A screenshot is below of what the Terminal looks after running start_windows.bat

A temporary fix is commenting out line 129 of tts-generation-webui/server.py but this removes the History tab.

Apparently this module doesn't get properly installed on Windows. Keep getting AttributeError regarding input method call when trying to save a generated audio file. Tried to install it manually using cmd, but error still occurs.

I encountered an error when trying to use my own quantizer. I have been testing this model, and it works, but I don't know where to add it in your code so that I can use my own. The one you provided only works in English. So, I placed it in the folder where the hubert models are and renamed it as "tokenizer," but I'm getting an error:

RuntimeError: Error(s) in loading state_dict for CustomTokenizer:
Unexpected key(s) in state_dict: "intermediate.weight", "intermediate.bias".
size mismatch for fc.weight: copying a param with shape torch.Size([10000, 4096]) from checkpoint, the shape in the current model is torch.Size([10000, 1024])

Using the code below, everything works fine.

import os
import torchaudio
import torch
import numpy as np

from hubert.hubert_manager import HuBERTManager
from hubert.pre_kmeans_hubert import CustomHubert
from hubert.customtokenizer import CustomTokenizer
from encodec import EncodecModel
from encodec.utils import convert_audio

hubert_model = CustomHubert(checkpoint_path='data/models/hubert/hubert.pt')

wav, sr = torchaudio.load('audio.wav')
if wav.shape[0] == 2:
    wav = wav.mean(0, keepdim=True)
semantic_vectors = hubert_model.forward(wav, input_sample_hz=sr)
tokenizer = CustomTokenizer.load_from_checkpoint('data/models/hubert/tokenizer.pth')
semantic_tokens = tokenizer.get_token(semantic_vectors)
model = EncodecModel.encodec_model_24khz()
model.set_target_bandwidth(6.0)
wav, sr = torchaudio.load('audio.wav')
wav = convert_audio(wav, sr, model.sample_rate, model.channels)
wav = wav.unsqueeze(0)
with torch.no_grad():
    encoded_frames = model.encode(wav)
codes = torch.cat([encoded[0] for encoded in encoded_frames], dim=-1).squeeze()
fine_prompt = codes
coarse_prompt = fine_prompt[:2, :]
np.savez('helloWorld.npz', semantic_prompt=semantic_tokens, fine_prompt=fine_prompt, coarse_prompt=coarse_prompt)

After generation this function is called
def save_cloned_voice( full_generation: FullGeneration, ): voice_name = f"test_clone_voice{str(np.random.randint(100000))}" filename = f"voices/{voice_name}.npz" save_npz(filename, full_generation) return filename

using the method save_npz from npz_tools.py
`def save_npz(filename: str, full_generation: FullGeneration, metadata: dict[str, Any]):
def pack_metadata(metadata: dict[str, Any]):
return list(json.dumps(metadata))

np.savez(
    filename,
    **{
        **compress_history(full_generation),
        "metadata": pack_metadata(metadata),
    },
)

`

Seems that metadata is missing. What can I use as Metadata?

The is a test call when calling npz_tools directly. That sets the metadata this way:

metadata_in = { "_version": "0.0.1", "_hash_version": "0.0.2", "_type": "bark", "is_big_semantic_model": True, "is_big_coarse_model": False, "is_big_fine_model": False, "prompt": "test", "language": None, "speaker_id": None, "hash": "98b14851692f09df5e89c68f0a8e2013", "history_prompt": "continued_generation", "history_prompt_npz": None, "history_hash": "98b14851692f09df5e89c68f0a8e2013", "text_temp": 0.7, "waveform_temp": 0.7, "date": "2023-06-07_16-56-09", "seed": "2039063546", }

Can simply copy paste this code to be used in save_cloned_voice? Or do I need to set the values like Language, speaker, prompt ...

When I put Length and Repetition values at a lower value, command line tells me I need to shorten my text, but I barely put any in.
Is there a way to fix this?

Hi, I faced this error when I tried to do a bark voice clone.

RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

I'm using a mac m2, and my device does not have proper GPU, so I unchecked "use GPT", but it still gets the issue.

Can you please help to check this??

Hi, I faced this issue when I tried to use RVC with mps (mac m2), and I got this issue.

没有发现支持的N卡, 使用MPS进行推理

logs...

RuntimeError: Input type (MPSHalfType) and weight type (torch.HalfTensor) should be the same

This is current GUI setting

Can you please check this issue?

When I generate anything, it works perfectly, however, sometimes I am finding I made a typo, mistake or something else and have to close it out entirely and come back to it again.

Can there be a halt or stop Generation when Generating Audio Samples?

Hey, i would like to run this somewhere in the cloud. Either on google colab, or a runpod.io container (that also has jupyter notebooks available).

Now the one click installer that is recommended, isn't very helpful in that case. I'd like to run this in a jupyter/colab notebook like people do with the other WebUIs out there. Is that an intended use case? Are there instructions floating around or do i have to adapt the manual installation instructions?

Regarding those manual installation: What do you mean with number 8: "clone the repos in the ./models/ directory and install requirements under them". I can't find a directory with that name. do you mean "./data/models/"? And what do i clone there? Can you please specify the repositories somewhere or add them as git submodules?

I think all the other instructions could simply be included in such a notebook. Thanks.

After executing the start_windows.bat, everything looked fine, but theres an error:

Loading Bark models
        - Text Generation:               GPU: Yes, Small Model: Yes
        - Coarse-to-Fine Inference:      GPU: Yes, Small Model: Yes
        - Fine-tuning:                   GPU: Yes, Small Model: No
        - Codec:                         GPU: Yes
Traceback (most recent call last):
  File "D:\**************\tts-generation-webui\tts-generation-webui\server.py", line 132, in <module>
    settings_tab_gradio(save_config_gradio, reload_config_and_restart_ui, gradio_interface_options)
  File "D:\**************\tts-generation-webui\tts-generation-webui\settings_tab_gradio.py", line 34, in settings_tab_gradio
    "inline": gr.Checkbox(label="inline: Display inline in an iframe", value=gradio_interface_options["inline"]),
KeyError: 'inline'

Done!
Drücken Sie eine beliebige Taste . . .

always appears when starting the start_windows.bat now How to solve ?

Hello;
Could this repo get a MIT or Apache2 license on it?

That could free people up to use and build upon it freely and comfortably.

Hey there, great project! For the newly added MusicGen section, can you add support for smaller gradations when it comes to Top-p and Temperature? Right now it can only go up by whole digits rather than tenths. I receive "Please enter a valid value" if I try to input something like 0.9 for Temperature.

config.json is missing

I have managed to get it up and running using Docker Install. However, while I can clone a voice in Bark, I am unable to render a TTS on my clone, or the preinstalled voice, either on Bark or TTS. The attached is a pic of the screen, which just hangs there...

I would really appreciate any pointers. Thanks

Here is a copy of the logs:

2023-08-01 18:01:47
2023-08-01 18:01:47 ==========
2023-08-01 18:01:47 == CUDA ==
2023-08-01 18:01:47 ==========
2023-08-01 18:01:47
2023-08-01 18:01:47 CUDA Version 11.8.0
2023-08-01 18:01:47
2023-08-01 18:01:47 Container image Copyright (c) 2016-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
2023-08-01 18:01:47
2023-08-01 18:01:47 This container image and its contents are governed by the NVIDIA Deep Learning Container License.
2023-08-01 18:01:47 By pulling and using the container, you accept the terms and conditions of this license:
2023-08-01 18:01:47 https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
2023-08-01 18:01:47
2023-08-01 18:01:47 A copy of this license is made available in this container at /NGC-DL-CONTAINER-LICENSE for your convenience.
2023-08-01 18:01:47
2023-08-01 18:01:58 Loading extensions:
2023-08-01 18:01:58 Loaded extension: callback_save_generation_musicgen_ffmpeg
2023-08-01 18:01:58 Loaded extension: callback_save_generation_ffmpeg
2023-08-01 18:01:58 Loaded extension: empty_extension
2023-08-01 18:01:58 Loaded 2 callback_save_generation extensions.
2023-08-01 18:01:58 Loaded 1 callback_save_generation_musicgen extensions.
2023-08-01 18:01:58 /app/tts-generation-webui/src/rvc_tab/rvc_tab.py
2023-08-01 18:01:58 Starting Gradio server...
2023-08-01 18:01:58 Gradio interface options:
2023-08-01 18:01:58 inline: False
2023-08-01 18:01:58 inbrowser: True
2023-08-01 18:01:58 share: False
2023-08-01 18:01:58 debug: False
2023-08-01 18:01:58 enable_queue: True
2023-08-01 18:01:58 max_threads: 40
2023-08-01 18:01:58 auth: None
2023-08-01 18:01:58 auth_message: None
2023-08-01 18:01:58 prevent_thread_lock: False
2023-08-01 18:01:58 show_error: False
2023-08-01 18:01:58 server_name: 0.0.0.0
2023-08-01 18:01:58 server_port: None
2023-08-01 18:01:58 show_tips: False
2023-08-01 18:01:58 height: 500
2023-08-01 18:01:58 width: 100%
2023-08-01 18:01:58 favicon_path: None
2023-08-01 18:01:58 ssl_keyfile: None
2023-08-01 18:01:58 ssl_certfile: None
2023-08-01 18:01:58 ssl_keyfile_password: None
2023-08-01 18:01:58 ssl_verify: True
2023-08-01 18:01:58 quiet: True
2023-08-01 18:01:58 show_api: True
2023-08-01 18:01:58 file_directories: None
2023-08-01 18:01:58 _frontend: True
2023-08-01 18:03:10 Running on local URL: http://0.0.0.0:7860
2023-08-01 18:03:10 gen old_generation_filename voices/voice_from_audio_1d36e051d27db45661738e095c0a8a9d.npz
2023-08-01 18:03:10 Loading Bark models
2023-08-01 18:03:10 - Text Generation: GPU: Yes, Small Model: Yes
2023-08-01 18:03:10 - Coarse-to-Fine Inference: GPU: Yes, Small Model: Yes
2023-08-01 18:03:10 - Fine-tuning: GPU: Yes, Small Model: Yes
2023-08-01 18:03:10 - Codec: GPU: Yes
Downloading text.pt: 14%|█▎ | 315M/2.32G [07:22<54:57, 607kB/s] /venv/lib/python3.10/site-packages/gradio/processing_utils.py:171: UserWarning: Trying to convert audio automatically from float32 to 16-bit int format.
2023-08-01 14:49:40 warnings.warn(warning.format(data.dtype))
Downloading text.pt: 19%|█▊ | 430M/2.32G [09:55<45:48, 686kB/s]
Downloading text_2.pt: 6%|▌ | 315M/5.35G [08:46<2:31:37, 554kB/s]

I have no experience with gradio otherwise i would even attempt it myself. I plan to even learn from it haha
My suggestion is to add a queue system to the voice generation. Instead of clicking generate, you would basically click add to queue. Process queue would then take care of it. This would help with line splitting as well. In this specific case i am talkign about tortoise. As i am getting way more stable results from that one

https://www.reddit.com/r/StableDiffusion/comments/zq0wl9/i_made_a_queue_system_for_automatic1111s_stable/
https://github.com/Kryptortio/SDAtom-WebUi-us
This is an example on how i envision it

when i press generate i am receiving an error
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/gradio/routes.py", line 437, in run_predict
output = await app.get_blocks().process_api(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1352, in process_api
result = await self.call_function(
File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1077, in call_function
prediction = await anyio.to_thread.run_sync(
File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 807, in run
result = context.run(func, *args)
File "/content/tts-generation-webui/src/rvc_tab/rvc_tab.py", line 72, in run_rvc
index_path=index_path.name,
AttributeError: 'NoneType' object has no attribute 'name'

@rsxdalv - Is there a way to save demuxed files?
On the video there are three dots on the right hand side of the separated track with a save option but I'm guessing this is a demo version. Presumably you will set up a button to feed the track to a vocoder or to a voice cloner.
(Helpful video by the way).

regards

@Magenta-6

So when I try to 'Use a voice' and it will come into an error and gives me this in the terminal:

Traceback (most recent call last):
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\routes.py", line 427, in run_predict output = await app.get_blocks().process_api(
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\blocks.py", line 1323, in process_api
result = await self.call_function(
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\blocks.py", line 1067, in call_function
prediction = await utils.async_iteration(iterator)
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\utils.py", line 336, in async_iteration
return await iterator.anext()
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\utils.py", line 329, in anext
return await anyio.to_thread.run_sync(
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 877, in run_sync_in_worker_thread
return await future
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\anyio_backends_asyncio.py", line 807, in run
result = context.run(func, *args)
File "C:\AI\one-click-installers-tts-4.0\installer_files\env\lib\site-packages\gradio\utils.py", line 312, in run_sync_iterator_async
return next(iterator)
File "C:\AI\one-click-installers-tts-4.0\tts-generation-webui\src\bark\generation_tab_bark.py", line 294, in gen
filename, filename_png, _, _, filename_npz, seed, metadata = generate(
File "C:\AI\one-click-installers-tts-4.0\tts-generation-webui\src\bark\generation_tab_bark.py", line 77, in generate
filename, filename_png, filename_npz, metadata = save_generation(
File "C:\AI\one-click-installers-tts-4.0\tts-generation-webui\src\bark\generation_tab_bark.py", line 130, in save_generation
history_hash = history_to_hash(history_prompt)
File "C:\AI\one-click-installers-tts-4.0\tts-generation-webui\src\bark\history_to_hash.py", line 11, in history_to_hash
"semantic_prompt": npz["semantic_prompt"].tolist(),
TypeError: string indices must be integers

Do you know how to fix this?

Okay I installed it and update still says that I need it. The installer took about 15 minutes and threw a million errors. After re-running it several times I finally get through it and it says "No GPU being used". Try to run the update bat and 15 minutes later it complains that I don't have MSVC. So I install that, try to update again and it still thinks I don't have it.

OS: Windows 11
GPU: GeForce 1650 (no Ti, anything like that)
Problem(s):

The very first error occured while installing the one-click-installer package itself:

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.
open-clip-torch 2.7.0 requires protobuf==3.20.0, but you have protobuf 4.23.4 which is incompatible.
rembg 2.0.35 requires scikit-image>=0.19.3, but you have scikit-image 0.19.2 which is incompatible.
tensorflow-intel 2.11.0 requires protobuf<3.20,>=3.9.2, but you have protobuf 4.23.4 which is incompatible.
tensorflow-intel 2.11.0 requires tensorboard<2.12,>=2.11, but you have tensorboard 2.13.0 which is incompatible.
audiocraft 0.0.2a2 requires hydra-core>=1.1, but you have hydra-core 1.0.7 which is incompatible.
Successfully installed Cython-0.29.36 Pillow-9.3.0 Werkzeug-2.3.6 absl-py-1.4.0 faiss-cpu-1.7.4 functorch-2.0.0 json5-0.9.14 librosa-0.9.2 llvmlite-0.39.0 matplotlib-inline-0.1.6 praat-parselmouth-0.4.3 protobuf-4.23.4 pyworld-0.3.3 resampy-0.4.2 rvc-beta-0.1.1 tensorboard-2.13.0 tensorboardX-2.6.1 torchcrepe-0.0.20 torchgen-0.0.1 tornado-6.3.2 traitlets-5.9.0 uvicorn-0.21.1

[notice] A new release of pip is available: 23.1.2 -> 23.2
[notice] To update, run: python.exe -m pip install --upgrade pip
Successfully installed RVC dependencies
Env file not found. Creating default env.
Config file not found. Creating default config.
Traceback (most recent call last):
  File "X:\tts_gen_webui\oobabooga_windows\installer_files\env\lib\site-packages\tensorboard\compat\__init__.py", line 42, in tf
    from tensorboard.compat import notf  # noqa: F401
ImportError: cannot import name 'notf' from 'tensorboard.compat' (X:\Bark\one-click-installers-tts-6.0\tts_gen_webui\oobabooga_windows\installer_files\env\lib\site-packages\tensorboard\compat\__init__.py)

After I tried to run it "anyway":

Traceback (most recent call last):
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\compat\__init__.py", line 42, in tf
    from tensorboard.compat import notf  # noqa: F401
ImportError: cannot import name 'notf' from 'tensorboard.compat' (X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\compat\__init__.py)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\utils\import_utils.py", line 1172, in _get_module
    return importlib.import_module("." + module_name, self.__name__)
  File "X:\tts_gen_webui\installer_files\env\lib\importlib\__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 1050, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
  File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 688, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 883, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\models\gpt2\modeling_gpt2.py", line 38, in <module>
    from ...modeling_utils import PreTrainedModel, SequenceSummary
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\modeling_utils.py", line 84, in <module>
    from accelerate import __version__ as accelerate_version
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\accelerate\__init__.py", line 3, in <module>
    from .accelerator import Accelerator
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\accelerate\accelerator.py", line 37, in <module>
    from .tracking import LOGGER_TYPE_TO_CLASS, GeneralTracker, filter_trackers
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\accelerate\tracking.py", line 42, in <module>
    from torch.utils import tensorboard
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\torch\utils\tensorboard\__init__.py", line 12, in <module>
    from .writer import FileWriter, SummaryWriter  # noqa: F401
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\torch\utils\tensorboard\writer.py", line 16, in <module>
    from ._embedding import (
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\torch\utils\tensorboard\_embedding.py", line 9, in <module>
    _HAS_GFILE_JOIN = hasattr(tf.io.gfile, "join")
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\lazy.py", line 65, in __getattr__
    return getattr(load_once(self), attr_name)
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\lazy.py", line 97, in wrapper
    cache[arg] = f(arg)
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\lazy.py", line 50, in load_once
    module = load_fn()
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tensorboard\compat\__init__.py", line 45, in tf
    import tensorflow
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\__init__.py", line 37, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\python\__init__.py", line 37, in <module>
    from tensorflow.python.eager import context
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\python\eager\context.py", line 28, in <module>
    from tensorflow.core.framework import function_pb2
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\core\framework\function_pb2.py", line 16, in <module>
    from tensorflow.core.framework import attr_value_pb2 as tensorflow_dot_core_dot_framework_dot_attr__value__pb2
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\core\framework\attr_value_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__pb2
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\core\framework\tensor_pb2.py", line 16, in <module>
    from tensorflow.core.framework import resource_handle_pb2 as tensorflow_dot_core_dot_framework_dot_resource__handle__pb2
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\core\framework\resource_handle_pb2.py", line 16, in <module>
    from tensorflow.core.framework import tensor_shape_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2
  File "C:\Users\Shiny\AppData\Roaming\Python\Python310\site-packages\tensorflow\core\framework\tensor_shape_pb2.py", line 36, in <module>
    _descriptor.FieldDescriptor(
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\google\protobuf\descriptor.py", line 561, in __new__
    _message.Message._CheckCalledFromGeneratedFile()
TypeError: Descriptors cannot not be created directly.
If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
If you cannot immediately regenerate your protos, some other possible workarounds are:
 1. Downgrade the protobuf package to 3.20.x or lower.
 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).

More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "X:\tts_gen_webui\tts-generation-webui\server.py", line 40, in <module>
    from src.tortoise.generation_tab_tortoise import generation_tab_tortoise
  File "X:\tts_gen_webui\tts-generation-webui\src\tortoise\generation_tab_tortoise.py", line 7, in <module>
    from src.tortoise.gen_tortoise import (
  File "X:\tts_gen_webui\tts-generation-webui\src\tortoise\gen_tortoise.py", line 7, in <module>
    from tortoise.api import TextToSpeech, MODELS_DIR
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tortoise\api.py", line 14, in <module>
    from tortoise.models.autoregressive import UnifiedVoice
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\tortoise\models\autoregressive.py", line 6, in <module>
    from transformers import GPT2Config, GPT2PreTrainedModel, LogitsProcessorList
  File "<frozen importlib._bootstrap>", line 1075, in _handle_fromlist
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\utils\import_utils.py", line 1163, in __getattr__
    value = getattr(module, name)
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\utils\import_utils.py", line 1162, in __getattr__
    module = self._get_module(self._class_to_module[name])
  File "X:\tts_gen_webui\installer_files\env\lib\site-packages\transformers\utils\import_utils.py", line 1174, in _get_module
    raise RuntimeError(
RuntimeError: Failed to import transformers.models.gpt2.modeling_gpt2 because of the following error (look up to see its traceback):
Descriptors cannot not be created directly.
If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
If you cannot immediately regenerate your protos, some other possible workarounds are:
 1. Downgrade the protobuf package to 3.20.x or lower.
 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).

More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates

RVC models one just drops into the GUI, but how do I select a custom Tortoise model and voice that I trained myself?

I cannot find a Tortoise folder or data/models folder (which I saw some folks here mention) in my installtion directory, nor can I find a dropdown or other thing in the GUI where I could select custom models. Though I did find a selection for Tortoise voices, I could not find out where those are installed to add my own.