Skip to content

A Python based project, which involves prediction autism in children using speech data using MFCC features (Training and Testing Data is Self Collected from NGOs and used with guardian permission)

Notifications You must be signed in to change notification settings

ronit1706/Autism-Detection

Repository files navigation

Autism-Detection

  • This project uses MFCC features, which are extracted from audio recordings, to train multiple models to predict autism.
  • The python file stores the trained models as a pkl file, which can later be used in predictor.py (CLI) or ui.py (Streamlit based UI).
  • The Models implemented are Random Forest, SVM, Naive Bayes and Artificial Neural Network.
  • The model that performs best is Random forest with a 90% accuracy, followed by Naive Bayes with 81% accuracy.

Working Model: Screenshot 2024-05-22 at 11 30 22 AM

Implementation: Screenshot 2024-05-22 at 11 23 54 AM Screenshot 2024-05-22 at 11 24 03 AM

About

A Python based project, which involves prediction autism in children using speech data using MFCC features (Training and Testing Data is Self Collected from NGOs and used with guardian permission)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages