windows ??

[zachysaur]

how to install on windows

[Nenesh]

If you don't want to use training, this should work (training don't work for me on windows without using WSL or Docker, "torchrun: command not found") :

git clone https://github.com/myshell-ai/MeloTTS.git cd MeloTTS conda create -n melotts python=3.10 conda activate melotts pip install -e . python -m unidic download

[zachysaur]

if i dont want to use docker and wsl

[zachysaur]

how i can train some voice?

[Nenesh]

Just install like i say, and test training. If training doesn't work, i don't know.

[arkilis]

No module named unidic

[YodaGitMaster]

Followed these steps:

git clone https://github.com/myshell-ai/MeloTTS.git
cd MeloTTS
conda create -n melotts python=3.10
conda activate melotts
pip install -e .
python -m unidic download

than used

melo-ui

Got error:

LookupError:
**********************************************************************
  Resource averaged_perceptron_tagger_eng not found.
  Please use the NLTK Downloader to obtain the resource:

  >>> import nltk
  >>> nltk.download('averaged_perceptron_tagger_eng')

  For more information see: https://www.nltk.org/data.html

  Attempted to load taggers/averaged_perceptron_tagger_eng/

  Searched in:
    - 'C:\\Users\\gamin/nltk_data'
    - 'C:\\Users\\gamin\\miniconda3\\envs\\melotts\\nltk_data'
    - 'C:\\Users\\gamin\\miniconda3\\envs\\melotts\\share\\nltk_data'
    - 'C:\\Users\\gamin\\miniconda3\\envs\\melotts\\lib\\nltk_data'
    - 'C:\\Users\\gamin\\AppData\\Roaming\\nltk_data'
    - 'C:\\nltk_data'
    - 'D:\\nltk_data'
    - 'E:\\nltk_data'
**********************************************************************

[louistiti]

@YodaGitMaster to fix this, you can download the NLTK data first:

Run the Python interpreter:

python

Import NLTK and download the dataset:

import nltk
nltk.download('averaged_perceptron_tagger_eng')

Or you can also try this directly:

python -m nltk.downloader averaged_perceptron_tagger_eng

If you still have some missing files, you can directly download all the data via by replacing averaged_perceptron_tagger_eng with all. But this will download many data that you might not need. You can also try with popular instead of all.

More info here: https://www.nltk.org/data.html.

Hope it helps! :smiley: