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

Voice emotion recognition model training method and electronic equipment

A technology for emotion recognition and model training, applied in speech recognition, speech analysis, instruments, etc., can solve the problem of low feasibility, and achieve the effect of good effect and balanced prediction

Pending Publication Date: 2021-11-02
EMOTIBOT TECH LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This approach is not very feasible

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
  • Voice emotion recognition model training method and electronic equipment
  • Voice emotion recognition model training method and electronic equipment
  • Voice emotion recognition model training method and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] The present invention will be further described below in conjunction with accompanying drawing.

[0052] At present, the corpus of speech emotion recognition is very scarce, mainly because of the difficulty and high cost of collecting speech emotion corpus. Generally, the collection of relevant speech and emotion corpus requires professional actors to record, and it is difficult for ordinary speakers to perform, so the collection is difficult, and the number of collections is difficult to reach a large scale. In addition, for some advanced emotional categories (such as: surprise, fear, disgust, contempt, doubt), the first officer has increased the difficulty of collection. Therefore, if a small amount of speech emotion corpus is used to train speech emotion recognition tasks, the recognition effect of the obtained speech emotion recognition model is limited, and it is difficult to obtain high accuracy.

[0053] In order to solve the above problems, a high-precision spe...

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 voice emotion recognition model training method and electronic equipment. The method comprises the following steps: acquiring a speaker recognition corpus; extracting frequency domain feature data from the speaker recognition corpus; training by using the frequency domain feature data to obtain a voice emotion feature extractor; acquiring a voice emotion corpus; extracting voice emotion feature data from the voice emotion corpus by using the voice emotion feature extractor; and training by using the voice emotion feature data to obtain a voice emotion recognition model. According to the method, the voice emotion recognition model obtained through training can also have relatively high accuracy only through a small amount of voice emotion corpora.

Description

technical field [0001] The invention relates to the technical field of speech emotion recognition, in particular to a speech emotion recognition model training method and electronic equipment. Background technique [0002] At present, the effect of deep learning in various fields is very good, which has the influence of improving the performance of computing hardware and deepening the model architecture. Among them, the size of the training corpus used is the most critical factor to achieve the above effects. Speech emotion recognition is a kind of deep learning that can be realized, but the corpus of speech emotion recognition is very scarce, and deep learning cannot be applied to achieve better recognition results. In contrast, the training corpus that speech recognition and speaker recognition can obtain is several thousand times, tens of thousands of times that of speech emotion recognition. [0003] To solve the problem of insufficient training data for speech emotion...

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): G10L15/06G10L25/63
CPCG10L15/063G10L25/63G10L2015/0631
Inventor 简仁贤许曜麒林长洲
Owner EMOTIBOT TECH LTD