Skip to content

README.md

Latest commit

 

History

History
31 lines (28 loc) · 1015 Bytes

README.md

File metadata and controls

31 lines (28 loc) · 1015 Bytes
Raw

Phoneme-Level-Labeling-Tool-GUI

This is a Phoneme-Level-Labeling-Tool developed using PyQt5 for annotating the dataset (.lab files generated from HSA) and boundary correcting the audio waveform with respect to its phonetic sounds. Please refer the GUI Demo video file provided to visually understand the tool's functionality.

Steps:

Installation

  1. Create a Python Virtual Environment [opt. Install Miniconda/Anaconda. Create a new conda virtual environment]
   python -m venv <env-name>
  1. Activate your new environment [Linux]
   source <env-name>/bin/activate
  1. Install the required packages
   pip install numpy pandas librosa sounddevice pyqt5 pyqtgraph

Executing the Code

  1. Run the labelingtool.py from terminal
    python3 labelingtool.py
  1. Load the .wav file and .lab file into the GUI with 'Load Audio' and 'Load Dataset' buttons
  2. Refer to the video demo of the GUI for understanding the flow and functionality of the GUI