Detection method for discriminating depression based on sound

Abstract

The invention discloses a method for discriminating depression based on voice discrimination. The method comprises the following steps: discriminating depression based on voice feature extraction anddeep learning processing; carrying out BSS algorithm analysis on the sound file data through the collection and storage of sound element datamation, and carrying out the recognition of voice; using MFCC as a characteristic parameter to analyze a voice signal to be processed, converting the voice signal to Mel frequency, and performing cepstrum analysis; adopting multiple groups of training data torespectively collect data in recording, and establishing a convolutional neural network model for discrimination; performing classification analysis on the obtained test sample data by using a BP neural network method; and adopting an ROC and AUC model evaluation method based on a confusion matrix to judge the accuracy of judging the depression suffering probability of an individual based on sound. The depression discrimination rate is obviously improved, and the cost is low.

Description

technical field [0001] The invention belongs to the technical field of speech processing, and more specifically relates to a method for detecting depression based on sound. Background technique [0002] Depression is a mental disorder accompanied by abnormal thinking and behavior, which has become a serious public health and social problem worldwide. According to a report published by the World Health Organization in 2017, more than 300 million people around the world are suffering from depression. In China, the number of patients with depression has reached 54 million (4.2% of the population), and the incidence rate is similar to the global level ( 4.4%); about 1.2 million young people aged 15-24 in China suffer from depression; the incidence of depression in Chinese college students is as high as 23.8% (similar to the data of British universities); UNICEF 2015 report It shows that the incidence of depression among adolescents in rural areas is higher than that of urban pe...

AI Technical Summary

Problems solved by technology

[0006] To sum up, the problems existing in the existing technology are: the traditional depression detection method is based on the SDS Depression Scale and the subjective judgment of clinicians, and there are large errors. Neural network algorithm binary classification and AUC accuracy verification, the detection method lacks scientific nature, and lacks effective objective evaluation indicators

Images