Unlock instant, AI-driven research and patent intelligence for your innovation.

A space stress emotion recognition method based on short-term memory weight fusion

A technology of emotion recognition and short-term memory, applied in speech analysis, instruments, etc., can solve problems such as Gaussian mixture model constraints, achieve timely early warning and monitoring, and improve the effect of recognition

Inactive Publication Date: 2016-08-03
SOUTHEAST UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the disadvantage of the Gaussian mixture model is also limited by the training data, and the successful establishment of an emotional model requires sufficient training data

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A space stress emotion recognition method based on short-term memory weight fusion
  • A space stress emotion recognition method based on short-term memory weight fusion
  • A space stress emotion recognition method based on short-term memory weight fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The technical solutions of the present invention will be further described below in conjunction with the drawings and embodiments.

[0035] figure 1 It is a block diagram of the speech emotion recognition system involved in the present invention, wherein the main modules include: segment feature extraction, emotion model training, frame-by-frame importance weight calculation, local emotion recognition results, global emotion vector fusion, etc. The implementation method of the system is introduced in detail below.

[0036] 1. Speech feature extraction

[0037] There are generally two construction methods for feature vectors for identification and modeling, static statistical features and short-term dynamic features. Dynamic features have a strong dependence on phoneme information. Changes in the text will have a greater impact on the emotional features. Emotional speech generally includes three sources of information, speaker information, semantic information and em...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a speech emotion recognition method based on short-term memory weight fusion under space stress environment. According to the law of short-term memory forgetting in psychology, the present invention extracts the prior probability of the recall accuracy rate of the speech frame; the prior probability of the recall accuracy rate, together with the adjustment coefficient, constitutes the importance weight of the frame; based on the importance of each frame Weight, calculate the judgment fusion weight of the speech segment; perform emotion recognition on each speech segment; perform emotion fusion according to the judgment fusion weight, and obtain the final recognition result. The method of the invention can effectively improve the recognition accuracy rate of the text-related speaker recognition system, and the system also has better robustness to noise.

Description

technical field [0001] The invention relates to a speech emotion recognition method, in particular to a speech emotion recognition method based on short-term memory weight fusion under space stress environment. Background technique [0002] In the special environment of aerospace, the emotional stability and objective evaluation of personnel are of great significance. In the case of closed space, weightless environment and long-term boring work, it is easy to induce stress speech emotions such as irritability. [0003] The traditional voice emotion automatic recognition technology mainly includes two levels: one is to use the characteristics of the voice signal as emotion recognition, that is, the problem of emotional feature extraction; the other is how to classify specific voice data, that is, pattern recognition The problem. [0004] Emotional expressions are not always obvious in speech signals. Different emotions can have similar characteristics, and different speake...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G10L25/63
Inventor 黄程韦王峻吴斌秦海波赵力魏昕王浩
Owner SOUTHEAST UNIV