Real-time-voice-cloning: Is there any way to make this work on macOS?

Created on 10 Mar 2020 · 3Comments · Source: CorentinJ/Real-Time-Voice-Cloning

This is as far as I got:

python3 ./demo_cli.py
WARNING:tensorflow:From /tmp/Real-Time-Voice-Cloning/synthesizer/models/modules.py:91: The name tf.nn.rnn_cell.RNNCell is deprecated. Please use tf.compat.v1.nn.rnn_cell.RNNCell instead.

Arguments:
    enc_model_fpath:   encoder/saved_models/pretrained.pt
    syn_model_dir:     synthesizer/saved_models/logs-pretrained
    voc_model_fpath:   vocoder/saved_models/pretrained/pretrained.pt
    low_mem:           False
    no_sound:          False

Running a test of your configuration...

Your PyTorch installation is not configured to use CUDA. If you have a GPU ready for deep learning, ensure that the drivers are properly installed, and that your CUDA version matches your PyTorch installation. CPU-only inference is currently not supported.

Given that I'm on macOS, and don't have an Nvidia GPU, how can I get this working?

Source

mrichman

👍1

Most helpful comment

I got this working on mac.
Use this fork instead https://github.com/shawwn/Real-Time-Voice-Cloning
In requirements.txt change the first line to tensorflow>=1.1.0,<=1.14.0
Run
pip install -r requirements.txt
pip install torch torchvision
You'll need to download and install the pretrained models before running python3 ./demo_cli.py or it'll complain that it can't find them but you'll be good to go.

monochromicorn on 15 Mar 2020

👍7

All 3 comments

monochromicorn on 15 Mar 2020

👍7

CPU-only inference is currently not supported ? i flow this ,then show me not support

GoToBoy on 2 May 2020

I've got great news for you, we recently added support for CPU inference (and even training)! It even works on macOS. Please pull the latest from the repo and try again.