Quantification method and system for emotional communication in voice social media

Abstract

The invention belongs to the technical field of information, and particularly discloses a method and system for calculating emotion spreading of the voice social contact media. The method includes the four steps of social contact media range voice preprocessing, voice characteristic parameter extracting, emotion recognition calculating and emotion spreading process quantizing. The system comprises four modules which have the functions of the four steps. According to the method and system, the emotion spreading of a social contact media range is calculated from the voice emotion perspective, and the emotion spreading process in the social contact media range is quantized and identified through a numerical value by means of PAD three-dimensional emotion parameter marking and BP neural network modeling. The method and system have the advantages that the characteristic that phrase type conversations are many in the voice social contact media is sufficiently considered, the voice characteristics are extracted at certain time intervals, the emotion recognition accuracy is easily improved, and emotions in the social contact media range can be dynamically reflected; an emotion spreading process quantizing graph is provided, and changes in the emotions in the social contact range can be conveniently observed visually.

Description

technical field [0001] The invention belongs to the field of information technology, and in particular relates to emotion computing technology, in particular to a method and system for quantifying emotion communication in voice social media. Background technique [0002] In recent years, with the rapid development of communication technology and the popularization of mobile smart terminals, new changes are taking place in the communication and communication methods of social media. Instant messaging tool software represented by QQ, WeChat, WhatsApp, etc. social media way. News about eating, drinking, playing, having fun, various aspects of life, and various positive and negative energies quickly spread to the general public through social media, a new means of communication. Since the emotions on social media often affect the emotions of individuals and groups in society, and even trigger group incidents, it is necessary to study the emotions of social media. At present, t...

AI Technical Summary

Problems solved by technology

[0003] 1. Due to the common phenomenon of polysemy in Chinese words, it has brought great difficulties to the construction of Chinese emotional dictionaries, and the emotional judgment based on network texts is completely dependent on the constructed Chinese emotional dictionaries, but the words marked in Chinese emotional dictionaries are ultimately Therefore, it is often impossible to avoid the phenomenon of emotion recognition errors and low recognition accuracy;

[0004] 2. Text-based sentiment analysis is based on probability statistics and matching of emotional words in sentences, and then gives a rough judgment of emotional tendency. The results obtained by this analysis cannot be quantified, and the results are not very intuitive

Images