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

Baby cry detection method and device

A detection method and crying technology, which is applied in the field of signal processing, can solve problems such as low separation accuracy, limited amount of data that can be processed, and inaccurate baby crying detection.

Inactive Publication Date: 2019-08-02
INST OF ACOUSTICS CHINESE ACAD OF SCI +1
View PDF9 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the defects existing in the voice signal baby cry detection technology at the present stage, including that the SVM learning algorithm has low separation accuracy for baby cry and other sounds, and the amount of data that can be processed is limited, which further causes the baby to cry In order to solve the problem that the sound detection is not accurate enough, a baby crying detection method and device are provided

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
  • Baby cry detection method and device
  • Baby cry detection method and device
  • Baby cry detection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The technical solutions provided by the present invention are further described below in conjunction with the examples.

[0047] Such as figure 1 as shown, figure 1 A schematic flow chart of a baby cry detection method provided for the implementation of the present invention, the method comprising:

[0048] Step 110 , performing perceptual linear prediction coefficient (PLP) feature extraction on the speech data in the pre-acquired sample training set to obtain speech features corresponding to the speech data in the sample training set.

[0049] The specific sample training set is established in advance by the outside world. For example, the user uses known samples to establish a sample training set in advance. Then, upload to the system. After the system obtains the sample training set, it performs PLP feature extraction on the voice data in the sample training set to obtain the voice features corresponding to the voice data in the sample selection line.

[0050] ...

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 relates to a baby cry detection method and device, and the method comprises the steps: carrying out the perception linear prediction coefficient PLP feature extraction of voice data in apre-obtained sample training set, and obtaining voice features corresponding to the voice data in the sample training set; training the voice data in the sample training set and the voice features corresponding to the voice data by using a deep neural network algorithm to obtain an acoustic model of baby cry; after PLP feature extraction is carried out on the test voice data, enabling the test voice data to be substituted into the acoustic model of baby cry for learning, and obtaining a posterior probability; and according to the posterior probability, determining whether the test voice datais voice data corresponding to the baby cry. The test voice data is classified by utilizing the acoustic model of the baby cry acquired in the mode, the voice type posterior probability correspondingto each frame of voice data in the voice data is acquired, and then whether the test voice data is the voice data corresponding to the baby cry or not is determined according to the voice type posterior probability, so that the test voice data is more accurate.

Description

technical field [0001] The invention relates to the technical field of signal processing, in particular to a method and device for detecting baby crying. Background technique [0002] Due to the similarities between baby crying and speech signals, more and more researchers use speech signal processing technology to perform preliminary processing on crying. Then the pattern recognition technology is used to judge the baby's cry after preliminary processing. The establishment of a baby cry recognition system requires three steps, which are preprocessing, feature extraction and classification algorithm establishment. The emphasis is on the extraction of feature parameters and the establishment of classification algorithms. Since infant crying and adult speech are produced by vocal organs with similar structures, algorithms similar to speech signal extraction and recognition can be used to extract and recognize infant crying. However, because there are other sounds in the bab...

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 Applications(China)
IPC IPC(8): G10L15/06G10L15/02
CPCG10L15/02G10L15/063G10L2015/0631
Inventor 葛凤培张少甫颜永红
Owner INST OF ACOUSTICS CHINESE ACAD OF SCI