Emotional music recommendation method based on brain-computer interaction

Abstract

The invention discloses an emotional music recommendation method based on brain-computer interaction. Music corresponding to emotions is automatically searched and recommended to a user by acquiring electroencephalogram signals of the user. The process includes the steps: firstly, extracting the EEG (electroencephalogram) signals of the user by an electroencephalogram acquisition instrument, performing wavelet decomposition on the EEG signals into four wave bands alpha, beta, gamma and delta, taking frequency band energy of the four wave bands as a feature, recognizing the emotions by a trained electroencephalogram emotion recognition model EMSVM, and judging emotion categories corresponding to the EEG signals; averagely decomposing external music signals into eight frequency bands within the range of 20Hz-20kHz, taking energy values of the eight frequency bands as characteristic values, recognizing music emotions by a trained music emotion recognition model MMSVM and building a music emotion database MMD; recommending the music corresponding to index numbers to the user according to the emotion categories of the electroencephalogram signals, and implementing an emotion-based music recommendation system. By the emotional music recommendation method, a new approach can be brought for infant music cultivation, sleep treatment and music search.

Description

technical field [0001] The invention relates to an emotional music recommendation method based on electroencephalogram signals, which can automatically recommend music according to the emotions by judging the current emotions of the users by analyzing the current electroencephalogram signals of the users. It is suitable for the fields of infant music training, sleep music therapy and general music search. Background technique [0002] Music exists in every corner of the world and has become one of the essential elements in people's life, study and spiritual therapy. The digital dissemination of music is becoming a popular trend. People have also begun to get used to getting rich and colorful music content from the Internet. Neurons in the cerebral cortex have spontaneous bioelectrical activity, which is recorded in the scalp as EEG signals. EEG signals have a corresponding relationship with people's emotional changes. The emotional state of a person can be judged by EEG ...

AI Technical Summary

Problems solved by technology

However, there is no method to use the user's physiological signals such as electroencephalogram (EEG) to automatically discover the user's emotion and then make recommendations.

Images