search

kuleshov / audio-super-res

Audio super resolution using neural networks
MIT License
1.18k stars 205 forks source link

Unclear instructions with pretrained model #55

Open chrisbward opened 1 year ago

chrisbward commented 1 year ago

So I've downloaded the pretrained model (thank you) and was just reading the instructions here;

https://github.com/kuleshov/audio-super-res#running-the-model

"Running the model" as follows;

Contents

The repository is structured as follows.

    ./src: model source code
    ./data: code to download the model data

Retrieving data

The ./data subfolder contains code for preparing the VCTK speech dataset. Make sure you have enough disk space and bandwidth (the dataset is over 18G, uncompressed). You need to type:

cd ./data/vctk;
make;

Next, you must prepare the dataset for training:

A little confused as I do not need to train the dataset now that I have the pretrained model?

Cheers!

Sawyerb commented 1 year ago

Sorry, the heading is a bit confusing. If you're using the pretrained model, you don't need to train on the dataset. That said, you still need to run the steps in the "Retrieving data" section to load in the audio files.