Voice emotion recognition model and method based on joint feature representation

Abstract

The invention discloses a voice emotion recognition model and method based on joint feature representation, and relates to voice emotion recognition technology. A convolutional recurrent neural network model is improved, a hidden layer in the neural network is configured to learn the joint feature representation of a spectral depth feature and a manual feature, and the joint feature extraction andsentiment classification are integrated into an end-to-end network model. The joint feature utilizes the complementarity between the spectral depth feature and the manual feature, makes full use of the emotional information carried in the voice, and more perfectly models the voice emotion. In addition, the end-to-end network model reduces parameter redundancy due to an intermediate output layer.The voice emotion recognition method based on joint feature representation improves the recognition accuracy of the voice emotions compared with an original voice emotion recognition method based on apure convolutional recurrent neural network.

Description

Technical field [0001] The present invention relates to speech emotion recognition technology, in particular to a speech emotion recognition model (HSF-CRNN) construction and speech emotion recognition method based on a convolutional recurrent neural network of joint feature representation. Background technique [0002] Emotion recognition helps to provide a humanized experience for human-computer interaction, so that computers can perceive the user's emotional state and analyze them, and then generate corresponding responses. It is an important ability necessary for computers in the future. Among them, voice is the basic way of human communication, and voice emotion recognition is particularly important. Speech emotion recognition is the process of calibrating emotion types for a given speech segment. Specifically, its task is to extract the acoustic features that can express emotions from the collected speech signals, and then map these features into certain types of emotions. ...

AI Technical Summary

Problems solved by technology

So far, neural network-based speech emotion recognition methods only learn emotional deep features from a single feature (such as spectral or hand-crafted features)

However, speech contains complex information, and various features can be extracted. Existing methods do not make full use of the complementarity between different features, which makes the modeling ability of speech emotion poor, resulting in relatively poor performance of emotion recognition. high

Images