Bridge the semantic gap between pop music acoustic feature and emotion: Build an interpretable model

Zhang, JiangLong; Huang, Xianglin; Yang, Lifang; Nie, Liqiang

doi:10.1016/j.neucom.2016.01.099

Cited by 16 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…15 A model for music emotion recognition using feature selection and statistical models has been proposed. 16 The results obtained had a higher average accuracy rate for arousal compared to valence (80% for arousal vs. 63% for valence). The combination of audio, lyrics, and linguistic data to classify Greek songs into several valence and arousal categories has been introduced.…”

Section: Related Workmentioning

confidence: 86%

Classification of Music Mood Using MPEG-7 Audio Features and SVM with Confidence Interval

Sarno

Ridoean

Sunaryono

et al. 2018

Int. J. Artif. Intell. Tools

View full text Add to dashboard Cite

Psychologically, music can affect human mood and influence human behavior. In this paper, a novel method for music mood classification is introduced. In the experiment, music mood classification was performed using feature extraction based on MPEG-7 features from the ISO/IEC 15938 standard for describing multimedia content. The result of this feature extraction are 17 low-level descriptors. Here, we used the Audio Power, Audio Harmonicity, and Audio Spectrum Projection features. Moreover, the discrete wavelet transform (DWT) was utilized for audio signal reconstruction. The reconstructed audio signals were classified by the new method, which uses a support vector machine with a confidence interval (SVM-CI). According to the experimental results, the success rate of the proposed method was satisfactory and SVM-CI outperformed the ordinary SVM.

show abstract

Section: Related Workmentioning

confidence: 86%

Classification of Music Mood Using MPEG-7 Audio Features and SVM with Confidence Interval

Sarno

Ridoean

Sunaryono

et al. 2018

Int. J. Artif. Intell. Tools

View full text Add to dashboard Cite

show abstract

“…It is difficult to induce an individual’s emotions effectively and measure the level of emotional arousal accurately [ 40 ]. The methods most used in emotion induction research were movie clips [ 41 ], personalized recall [ 42 ], picture viewing [ 43 ], acoustic material [ 44 ] and standardized imagery [ 45 ]. Analysis results showed that music, movies and imagery were relatively ideal methods for emotion induction and the success rate was more than 75% [ 46 , 47 ].…”

Section: Study 2-driving Intention Prediction Models Adapting To Mmentioning

confidence: 99%

Differences in Driving Intention Transitions Caused by Driver’s Emotion Evolutions

Liu

Wang

2020

IJERPH

View full text Add to dashboard Cite

Joining worldwide efforts to understand the relationship between driving emotion and behavior, the current study aimed at examining the influence of emotions on driving intention transition. In Study 1, taking a car-following scene as an example, we designed the driving experiments to obtain the driving data in drivers’ natural states, and a driving intention prediction model was constructed based on the HMM. Then, we analyzed the probability distribution and transition probability of driving intentions. In Study 2, we designed a series of emotion-induction experiments for eight typical driving emotions, and the drivers with induced emotion participated in the driving experiments similar to Study 1. Then, we obtained the driving data of the drivers in eight typical emotional states, and the driving intention prediction models adapted to the driver’s different emotional states were constructed based on the HMM severally. Finally, we analyzed the probabilistic differences of driving intention in divers’ natural states and different emotional states, and the findings showed the changing law of driving intention probability distribution and transfer probability caused by emotion evolution. The findings of this study can promote the development of driving behavior prediction technology and an active safety early warning system.

show abstract

“…Many current methods for obtaining insights into deep network-based audio classification systems do not explain the predictions in a human understandable way but rather design special filters that can be visualized [13], or analyze neuron activations [8]. To the best of our knowledge, [19] is the only attempt to build an interpretable model for MER. They performed the task of feature extraction and selection and built models from different model classes on top of them.…”

Section: Related Workmentioning

confidence: 99%

Towards Explainable Music Emotion Recognition: The Route via Mid-level Features

Chowdhury¹,

Vall²,

Haunschmid³

et al. 2019

Preprint

View full text Add to dashboard Cite

Emotional aspects play an important part in our interaction with music. However, modelling these aspects in MIR systems have been notoriously challenging since emotion is an inherently abstract and subjective experience, thus making it difficult to quantify or predict in the first place, and to make sense of the predictions in the next. In an attempt to create a model that can give a musically meaningful and intuitive explanation for its predictions, we propose a VGG-style deep neural network that learns to predict emotional characteristics of a musical piece together with (and based on) human-interpretable, mid-level perceptual features. We compare this to predicting emotion directly with an identical network that does not take into account the mid-level features and observe that the loss in predictive performance of going through the mid-level features is surprisingly low, on average. The design of our network allows us to visualize the effects of perceptual features on individual emotion predictions, and we argue that the small loss in performance in going through the midlevel features is justified by the gain in explainability of the predictions.

show abstract

Bridge the semantic gap between pop music acoustic feature and emotion: Build an interpretable model

Cited by 16 publications

References 46 publications

Classification of Music Mood Using MPEG-7 Audio Features and SVM with Confidence Interval

Classification of Music Mood Using MPEG-7 Audio Features and SVM with Confidence Interval

Differences in Driving Intention Transitions Caused by Driver’s Emotion Evolutions

Towards Explainable Music Emotion Recognition: The Route via Mid-level Features

Contact Info

Product

Resources

About