Vocal-Data-Analysis

Analysing human voice to predict their age, gender and accent

Getting Started

To setup the environment and get the code running, follow the steps below.

Software Used

Major Frameworks and Libraries that this project is built using.

• Python

• Librosa

• Scikit-Learn

• Jupyter Notebook

• SVM

• KNN

• List item

• Pandas

Prerequisites

This is a list of things you need to use the software, and how to install them.

• Python3

  1. Visit their official site and download the package for your OS.

  2. Install it.

• Anaconda

  1. Visit Anacondas official site and follow the installation process.

  2. To verify that conda has been install, Run ->

  3. conda --version

• Librosa and ffmeg

  1. Run ->

     conda install -c conda-forge ffmpeg

     conda install librosa

     pip install glob

     pip install sklearn

     pip install csv

     pip install python_speech_features

Instructions to execute the code

• Inside terminal navigate to the location of the code. Now run the command ->

  1. jupyter notebook

  2. For training data extraction purpose, use the file ‘voice.ipynb’

  3. For looking at the data in form of graphs use file ‘voice-process.ipynb’

  4. For training and testing the model use the file ‘voice-pred.ipynb’

  5. Make sure to run the import part of the code first.

  6. Now you can jump to the part of the code that u want to train.