Related Experiment Video
Updated: Jan 10, 2026
05:19
Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
Published on: July 7, 2023
3.2K
Utilizing Temporal Inductive Path Neural Networks for Accurate Voice-Based Depression Classification: A Detailed
K Ashok Kumar1, Narsaiah Domala2, Vijaya Kumar Sajjan3
1Department of Electronics and Communication Engineering, Bhoj Reddy Engineering College for Women, Hyderabad, Telangana.
Journal of Voice : Official Journal of the Voice Foundation
|November 20, 2025
Summary
This study introduces a voice-based depression classification system (AVA-TIPNN-DD) using temporal inductive path neural networks. The novel method accurately detects depression from speech patterns, offering a promising solution for mental health assessment.
Area of Science:
- Artificial Intelligence
- Computational Linguistics
- Psychiatry
Background:
- Depressive disorder is a prevalent mental illness impacting mood and behavior, often leading to significant life difficulties.
- Hesitancy to seek psychiatric care due to long waiting times and high costs presents a barrier to treatment.
- Accurate and accessible diagnostic tools are crucial for timely intervention and management of depression.
Purpose of the Study:
- To propose a novel voice-based depression classification system named AVA-TIPNN-DD.
- To leverage temporal inductive path neural networks (TIPNNs) for analyzing speech patterns to identify mental health states.
- To enhance TIPNN performance through optimization using the binary battle royale algorithm.
Main Methods:
- Voice recordings were collected and preprocessed using the Koopman Kalman particle filter for audio cleaning.
- Spectrum features (power spectrum density, spectral power, spectral centroid, spectral flatness measure) were extracted using multi-objective matched synchro squeezing chirplet transform.
- The extracted features were fed into an optimized TIPNN (using binary battle royale algorithm) for depression classification.
Main Results:
- The AVA-TIPNN-DD method achieved high performance metrics: 99.26% Accuracy, 99.5% Precision, 99.6% Sensitivity, and 99.5% F1-Score.
- The proposed system demonstrated superior performance compared to existing methods like DSS-DD-CNN, TBFA-DD-FFANN, and SVR-DD-DRN.
- The binary battle royale optimizer effectively improved TIPNN's parameter optimization for accurate depression detection.
Conclusions:
- The AVA-TIPNN-DD system offers a highly accurate and efficient voice-based approach for depression classification.
- This method addresses accessibility issues in mental healthcare by providing a potentially low-cost and timely diagnostic tool.
- Voice analysis holds significant potential for the early detection and management of depressive disorders.
