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

A Lie Detection Method Based on Fixed-duration Speech Emotion Recognition Sequence Analysis

A technology of speech emotion recognition and sequence analysis, which is applied in speech analysis, sensors, diagnostic recording/measurement, etc., and can solve problems such as late start of research

Active Publication Date: 2021-09-24
SOUTHEAST UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The research on polygraph technology focusing on speech features as clues started relatively late, and most people pay attention to the influence of acoustic features on speech polygraph detection, but so far, no single feature can be used directly and effectively for polygraph detection

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 Lie Detection Method Based on Fixed-duration Speech Emotion Recognition Sequence Analysis
  • A Lie Detection Method Based on Fixed-duration Speech Emotion Recognition Sequence Analysis
  • A Lie Detection Method Based on Fixed-duration Speech Emotion Recognition Sequence Analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] Embodiment 1: as figure 1 As shown, the present invention proposes a kind of polygraph method based on fixed duration speech emotion recognition sequence analysis, and the detailed steps of this method are:

[0078] Step 1: Establish a Chinese polygraph corpus and process it;

[0079] Step 1.1: Establish a Chinese lie detection corpus;

[0080] Step 1.2: Propose the truth fragments in the lie data;

[0081] Step 1.3: Divide the truth and lies into corpus segments of equal duration, and label them to facilitate subsequent experiments;

[0082] Step 2: Preprocessing the speech;

[0083] Step 2.1: discretize the speech signal, and use a first-order high-pass filter for pre-emphasis. The expression of the first-order high-pass filter is as follows:

[0084] H(z)=1-αz -1 ,0.9<α<1.0;

[0085] Step 2.2: Framing the signal, the frame length is 30ms, and the frame shift is 10ms;

[0086] Step 2.3: Select the Hamming window function, whose calculation formula is as follows...

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 polygraph method based on fixed-duration voice emotion recognition sequence analysis. The main content of the method is: firstly, the recorded polygraph corpus is processed to make it into two types of short-term corpora of equal length, which is convenient for subsequent experiments ; Then perform pre-processing work such as pre-emphasis, framing and windowing on the speech; based on the time-frequency characteristics of the speech, extract including pitch frequency, MFCC, formant, short-term energy and short-term average zero-crossing rate and its statistical features, etc. Speech emotion features; use decision tree for feature selection, and finally form feature vectors; use SVM to train the corpus, predict the tested voice, and output the voice emotion results regularly; use the timing output voice emotion results for polygraph analysis. The present invention adopts the method of decision tree to select features, and obtains higher accuracy; outputs the result in the form of vector, and fully considers the emotional changes in the process of lying.

Description

technical field [0001] The invention belongs to non-contact polygraph detection, in particular to a polygraph method based on fixed-duration speech emotion recognition sequence analysis, and belongs to the technical field of polygraph detection. Background technique [0002] Voice is the most direct and convenient way for people to communicate. The polygraph based on voice is non-contact. The recording equipment is simple and does not require very complicated equipment. The preparation time is very short, and the subjects will not have great psychological pressure. Thereby improving the accuracy of the analysis. This is the significance of the research carried out in this paper. Speech contains a lot of information about the speaker, such as the speaker's identity, gender and age, and even personality. Early studies have shown that speech contains the speaker's emotional state, implying many reliable relationships between speech features and specific emotions. When people...

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): G10L25/63G10L25/48G10L25/27G10L25/24G10L25/21A61B5/16A61B5/00
CPCA61B5/164A61B5/4803A61B5/7246G10L25/21G10L25/24G10L25/27G10L25/48G10L25/63
Inventor 李玉峰黄永明
Owner SOUTHEAST UNIV