DIS-NV feature-based emotion recognition method

Abstract

The invention discloses a DIS-NV feature-based emotion recognition method. The method comprises the following steps of S1, selecting an emotion word having the non-fluency feature as a DIS feature word, selecting a sound production type having the non-language feature as an NV feature class, and constructing a DIS-NV feature set; S2, obtaining training voice texts, respectively matching the training voice texts with the DIS-NV feature set, extracting corresponding DIS-NV feature values, training by adopting a BLSTM model and obtaining a BLSTM classification model; S3, obtaining a to-be-recognized voice text, matching the to-be-recognized voice text with the DIS-NV feature set, extracting a corresponding DIS-NV feature value, recognizing by using the BLSTM classification model and outputting an emotion recognition result. According to the invention, the emotion recognition is realized through fully utilizing voice texts having the non-fluency feature and the non-language feature. The method is simple in implementation, high in recognition efficiency and high in precision.

Description

technical field [0001] The invention relates to the technical field of automatic emotion recognition, in particular to an emotion recognition method based on DIS-NV features. Background technique [0002] Emotion recognition is to identify the emotional information in the dialogue to judge the emotional state of the speaker. Through the automatic recognition of emotion, better human-computer interaction can be achieved, such as human-computer communication, conversational agency, etc. At present, emotion recognition is usually based on a category of processing methods, that is, there are several basic and common emotions in the brain, such as: happiness, sadness, surprise, fear, anger, and disgust, but the speaker Emotional states are usually complex, and a single emotional expression or limited discrete categories are usually difficult to properly describe the complex emotional state. One solution is to replace the classification with continuous emotional labels in a multid...

AI Technical Summary

Problems solved by technology

The number of LLD features is very large, basically between 1000 and 2000. On the one hand, the training of emotion recognition models for a large number of feature values ​​is difficult and requires a long training time, making the recognition efficiency low; on the other hand, the number of features Too much will also bring a lot of information redundancy to a certain extent, and it is difficult to effectively control the dependencies between features, thus affecting the accuracy of the trained emotion recognition model

Images