Skip to content

onkyo14taro/auto-grbas

Repository files navigation

Automatic GRBAS Scoring of Pathological Voices using Deep Learning and a Small Set of Labeled Voice Data

About

This repository contains the implementation of "Automatic GRBAS Scoring of Pathological Voices using Deep Learning and a Small Set of Labeled Voice Data".

Note that some scripts in Supplementary_Programs/ are only available in "Supplementary Materials" of our paper.

Contents

  • data/: Dummy dataset (all audio data were identical and uttered by the author, Hidaka).
  • notebooks/: Examples of time-frequency representations and data augmentation techniques.
  • sample_audio/: Sample audio files used in the notebooks.
  • shell/: Shell scripts.
  • src/: Python scripts.
  • tests/: Test scripts.
  • Supplementary_Programs/: Programs for acoustic feature calculation.

Dependencies

Install Poetry if it is not already installed.

pip install poetry

You can install the necessary packages with the following command:

poetry install

Usage

First, activate the virtual environment created by Poetry.
Then, you can run an experiment on the dummy dataset with the following code:

./shell/sample.sh

Note that the dataset is dummy, so the learning will not be successful.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published