Infant crying emotion recognition method based on improved long-short-time memory network

A long-short-term memory and emotion recognition technology, applied in the field of emotion recognition, can solve the problems of low efficiency of classification algorithms and different lengths of speech, achieve remarkable recognition performance, high recognition efficiency, and overcome the effect of unequal length of speech samples

Active Publication Date: 2019-01-18
NANJING INST OF TECH
View PDF10 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to overcome the problems of different speech lengths

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
  • Infant crying emotion recognition method based on improved long-short-time memory network
  • Infant crying emotion recognition method based on improved long-short-time memory network
  • Infant crying emotion recognition method based on improved long-short-time memory network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] Such as figure 1 As shown, the baby crying emotion recognition method based on the improved long-short-term memory network of the present invention comprises the following steps,

[0040]Step (A), carrying out endpoint detection and frame division of the voice of the baby crying data set, extracting the timing-related features of the voice of the baby crying data set, including the following steps,

[0041] (A1), the voice of the baby crying data set is subjected to endpoint detection, which is used to eliminate the silent segment to ensure that the timing-related features are extracted from the effective voice information;

[0042] (A2), the voice of the baby crying data set after the endpoint detection is divided into frames according to a frame of every 40ms, and is divided into multiple groups of frame data;

[0043] (A3), each group of multi-group frame d...

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 an infant crying emotion recognition method based on an improved long-short-time memory network. The infant crying data set speech is detected by endpoints and divided into frames, the temporal correlation characteristics of the infant crying data set speech are extracted, and a processing algorithm of the long-short time memory network is established according to the temporal correlation characteristics of different lengths. Then, the strategy of combining attention mechanism with depth of temporal sequence is introduced into the forgetting gate of the long-short memory network. The experimental results show that this method can not only greatly reduce the model parameters, but also show significant recognition performance on the recorded infant emotion database, and the recognition efficiency is high, which has a good application prospect.

Description

technical field [0001] The invention relates to the technical field of emotion recognition, in particular to a baby cry emotion recognition method based on an improved long-short-term memory network. Background technique [0002] The cry of a newborn baby is a way of expression for a child, and it is extremely important for parents to know their baby's needs in a timely manner. The sense of security gained in the first few months of life will accompany the baby's future life, so meeting the baby's needs in time will make the baby grow up healthier. For a "croaking" baby on the ground, there are limited ways to express emotional needs information. Basically, it can only express needs and moods by crying and laughing. However, babies usually express their mood at that time by laughing, and the information it conveys is relatively simple and clear, generally expressing psychological and emotional states such as happiness and joy. Therefore, babies can only express their needs...

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
IPC IPC(8): G10L25/63G10L25/27G10L25/87G10L25/03
CPCG10L25/03G10L25/27G10L25/63G10L25/87
Inventor 梁瑞宇梁镇麟谢跃赵力唐闺臣
Owner NANJING INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products