A Survey on Human Emotion Recognition Approaches,  Databases and Applications

Vinola, C; Vimaladevi, K.

doi:10.5565/rev/elcvia.795

Cited by 62 publications

(31 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Emotion recognition from the para-lingual component of the speech has been an active research area for decades [3][4][5][6][7][8]. Traditional methods were based on short-time frame-level feature extraction, followed by utterance-level information extraction, and classification or regression as required [3][4][5][6][7][8]. In the recent years, deep learning methodologies and tools have been introduced to this area, used for feature extraction, classification/regression, or both [9][10][11][12][13][14].…”

Section: Related Work and Evaluation Datasetmentioning

confidence: 99%

“…These advanced solutions capture and analyze the human emotions from multiple sources, where the human voice serves as one of the major ones. Emotion recognition capabilities are required to support natural and efficient human-computer interaction, generate marketing insights, help discovering entertainment content across large repositories and playlists, enable effective elearning through e-tutors, and many more [1][2][3]. Emotion recognition from the para-lingual information in the speech has gone through a significant improvement over the recent years, with the introduction of deep neural networks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Rozenberg²,

2017

View full text Add to dashboard Cite

We present a new implementation of emotion recognition from the para-lingual information in the speech, based on a deep neural network, applied directly to spectrograms. This new method achieves higher recognition accuracy compared to previously published results, while also limiting the latency. It processes the speech input in smaller segments-up to 3 seconds, and splits a longer input into non-overlapping parts to reduce the prediction latency. The deep network comprises common neural network toolsconvolutional and recurrent networks-which are shown to effectively learn the information that represents emotions directly from spectrograms. Convolution-only lowercomplexity deep network achieves a prediction accuracy of 66% over four emotions (tested on IEMOCAP-a common evaluation corpus), while a combined convolution-LSTM higher-complexity model achieves 68%. The use of spectrograms in the role of speech-representing features enables effective handling of background non-speech signals such as music (excl. singing) and crowd noise, even at noise levels comparable with the speech signal levels. Using harmonic modeling to remove non-speech components from the spectrogram, we demonstrate significant improvement of the emotion recognition accuracy in the presence of unknown background non-speech signals.

show abstract

Section: Related Work and Evaluation Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Rozenberg²,

2017

View full text Add to dashboard Cite

show abstract

“…This shows that， α is an eigenvector of y x K K corresponding to the eigenvalue 2 σ . From the first equation of Equation (8) and 0…”

Section: Incomplete Cholesky Decomposition Based Kernel Cross Modal Fmentioning

confidence: 99%

Incomplete Cholesky Decomposition based Kernel Cross Modal Factor Analysis for Audiovisual Continuous Dimensional Emotion Recognition

2019

KSII TIIS

View full text Add to dashboard Cite

Recently, continuous dimensional emotion recognition from audiovisual clues has attracted increasing attention in both theory and in practice. The large amount of data involved in the recognition processing decreases the efficiency of most bimodal information fusion algorithms. A novel algorithm, namely the incomplete Cholesky decomposition based kernel cross factor analysis (ICDKCFA), is presented and employed for continuous dimensional audiovisual emotion recognition, in this paper. After the ICDKCFA feature transformation, two basic fusion strategies, namely feature-level fusion and decision-level fusion, are explored to combine the transformed visual and audio features for emotion recognition. Finally, extensive experiments are conducted to evaluate the ICDKCFA approach on the AVEC 2016 Multimodal Affect Recognition Sub-Challenge dataset. The experimental results show that the ICDKCFA method has a higher speed than the original kernel cross factor analysis with the comparable performance. Moreover, the ICDKCFA method achieves a better performance than other common information fusion methods, such as the Canonical correlation analysis, kernel canonical correlation analysis and cross-modal factor analysis based fusion methods.

show abstract

“…Following [6,23], we group valence, activation and dominance into three categories: low: [1,2]; medium: (2,4); high: [4,5]. The gender is treated as a two-way classification task.…”

Section: Experiments 2: Multi-task Learningmentioning

confidence: 99%

“…Speech emotion recognition, as an important aspect of emotion recognition, has changed deeply under the influence of deep learning (DL) [1][2][3]. The traditional method is a multi-step process [4,5]. Firstly, original signals are divided into overlapping frames and frame-level features are extracted.…”

Section: Introductionmentioning

confidence: 99%

Speech Emotion Recognition via Contrastive Loss under Siamese Networks

Lian

Tao

et al. 2018

Proceedings of the Joint Workshop of the 4th Workshop on Affective Social Multimedia Computing and First Multi-Modal Affective

View full text Add to dashboard Cite

Speech emotion recognition is an important aspect of humancomputer interaction. Prior work proposes various end-to-end models to improve the classification performance. However, most of them rely on the cross-entropy loss together with softmax as the supervision component, which does not explicitly encourage discriminative learning of features. In this paper, we introduce the contrastive loss function to encourage intra-class compactness and inter-class separability between learnable features. Furthermore, multiple feature selection methods and pairwise sample selection methods are evaluated. To verify the performance of the proposed system, we conduct experiments on The Interactive Emotional Dyadic Motion Capture (IEMOCAP) database -a common evaluation corpus. Experimental results reveal the advantages of the proposed method, which reaches 62.19% in the weighted accuracy and 63.21% in the unweighted accuracy. It outperforms the baseline system that is optimized without the contrastive loss function with 1.14% and 2.55% in the weighted accuracy and the unweighted accuracy, respectively.

show abstract

A Survey on Human Emotion Recognition Approaches, Databases and Applications

Cited by 62 publications

References 82 publications

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Efficient Emotion Recognition from Speech Using Deep Learning on Spectrograms

Incomplete Cholesky Decomposition based Kernel Cross Modal Factor Analysis for Audiovisual Continuous Dimensional Emotion Recognition

Speech Emotion Recognition via Contrastive Loss under Siamese Networks

Contact Info

Product

Resources

About