nii-yamagishilab / ZMM-TTS

ZMM-TTS: Zero-shot Multilingual and Multispeaker Speech Synthesis Conditioned on Self-supervised Discrete Speech Representations
BSD 3-Clause "New" or "Revised" License
110 stars 8 forks source link

packages not available #2

Open kunibald413 opened 6 months ago

kunibald413 commented 6 months ago

using conda to create with the requirements.txt file gives this error:

Solving environment: unsuccessful attempt using repodata from current_repodata.json, retrying with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed

and lists all lines in the requirements.txt file for packages that it can't find.

doing a cross check, I was able to install numba in conda with this conda install numba

could you provide the channels, or a requirements file in pip format?

gongchenghhu commented 6 months ago

@igbeller Thanks for trying. If you cannot create an environment with conda, you can try the following instructions:

conda create --name ZMM-TTS  python=3.8
conda activate ZMM-TTS 

conda install  blas=1.0=mkl  pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.3  pytorch-mutex==1.0=cuda  libgcc-ng=11.2.0=h1234567_1 libgfortran-ng=11.2.0=h00389a5_1 libgfortran5=11.2.0=h1234567_1 numpy-base==1.23.5  -c conda-forge -c pytorch

pip install speechbrain==0.5.14 text2phonemesequence==0.1.4 matplotlib==3.6.2 g2p  pandas==1.5.2 g2p-en==2.1.0 epitran==1.24 praat-parselmouth==0.4.3 pyworld==0.3.2 pycwt==0.3.0a22 python-speech-features==0.6 pstring==1.0.2

pip install transformers==4.30.2 keras==2.13.1 librosa==0.9.2   # mkl==2021.4.0

pip install tensorflow-gpu==2.2.0 unidecode==1.2.0 pypinyin==0.48.0

pip install  tensorboardx==2.6 tensorboard==2.2.2

If you are still facing any issues, please feel free to ask.