Detection of Human Emotion from Speech—Tools and Techniques

Mohanta, Abhijit; Sharma, Uzzal

doi:10.1007/978-981-10-6626-9_20

Cited by 3 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The implementation of ReLu is as follows: if the input number is negative, then set it to zero (when the input is less than zero), or select the same number (when the number is more than zero). This function makes faster to train networks as in formula (2).…”

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

“…One of these applications is human emotion recognition, which is considered as an attempt for the computing system to understand human face emotion. Generally, human emotion can be carried out via speech emotion [2] or face emotion systems [3,4]. Face emotion is the closest to shape recognition, as it is considered important in such applications of computer vision, entertainment, and robotics [5] and remains the focus of this paper.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Smiling and Non-smiling Emotion Recognition Based on Lower-half Face using Deep-Learning as Convolutional Neural Network

Malallah¹,

Al-Jubouri²,

Sabaawi³

2020

Proceedings of the Proceedings of the 1st International Multi-Disciplinary Conference Theme: Sustainable Development and Smart

View full text Add to dashboard Cite

Image understanding is considered important among researchers. In this paper, a new technique is proposed to classify a detected face into two classes as a smile or nonsmile category. First, the system detects and segments only the face. Then, it converts the image from RGB to Gray-Scale and enhances the image via an equalization technique. Where the contribution of this research is depending only on the lower half of the face since most of the smiling information can be perceived from the mouth and its perimeter. Then, a convolutional neural network (CNN) is applied to generate two output nodes. Public GENKI-4K database is used for the experiments, which contains 4000 challenged face images. The results demonstrate that the accuracy with 4-Fold cross-validation is 91%. This approach achieves a promising performance as compared with the state-of-the-art techniques in both accuracy and processing time.

show abstract

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Smiling and Non-smiling Emotion Recognition Based on Lower-half Face using Deep-Learning as Convolutional Neural Network

Malallah¹,

Al-Jubouri²,

Sabaawi³

2020

Proceedings of the Proceedings of the 1st International Multi-Disciplinary Conference Theme: Sustainable Development and Smart

View full text Add to dashboard Cite

show abstract

“…Emotion recognition plays a vital role in human-machine interactions, which can improve the level of user experience and provide great interactivity [1]. With the rapid growth of human-computer interaction, speech emotion recognition (SER) has attracted steadily increasing attention from researchers.…”

Section: Introductionmentioning

confidence: 99%

Representation Learning with Spectro-Temporal-Channel Attention for Speech Emotion Recognition

Guo

Wang

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Convolutional neural network (CNN) is found to be effective in learning representation for speech emotion recognition. CNNs do not explicitly model the associations or relative importance of features in the spectral/temporal/channelwise axes. In this paper, we propose an attention module, named spectro-temporal-channel (STC) attention module that is integrated with CNN to improve representation learning ability. Our module infers an attention map along the three dimensions, namely time, frequency, and CNN channel. Experiments are conducted on the IEMOCAP database to evaluate the effectiveness of the proposed representation learning method. The results demonstrate that the proposed method outperforms the traditional CNN method by an absolute increase of 3.13% in terms of F1 score.

show abstract

“…An utterance may contain subject dependent auditory clues regarding expressed emotions which are not captured through speech transcripts alone. With deep learning, architectures can extract 1 Note: this has been updated from the ICASSP published version due to a small error. Results have been updated to correct the error, but overall findings are unchanged.…”

Section: Introductionmentioning

confidence: 99%

“…With the advance of technology, Human Computer Interaction (HCI) has become a major research area. Within this field, automatic emotion recognition is being pursued as a means to improve the level of user experience by tailoring responses to the emotional context, especially in humanmachine interactions [1]. However, this remains challenging due to the ambiguity of expressed emotions.…”

Section: Introductionmentioning

confidence: 99%

Attention Driven Fusion for Multi-Modal Emotion Recognition

Priyasad

Fernando

Denman

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Deep learning has emerged as a powerful alternative to hand-crafted methods for emotion recognition on combined acoustic and text modalities. Baseline systems model emotion information in text and acoustic modes independently using Deep Convolutional Neural Networks (DCNN) and Recurrent Neural Networks (RNN), followed by applying attention, fusion, and classification. In this paper, we present a deep learning-based approach to exploit and fuse text and acoustic data for emotion classification. We utilize a SincNet layer, based on parameterized sinc functions with band-pass filters, to extract acoustic features from raw audio followed by a DCNN. This approach learns filter banks tuned for emotion recognition and provides more effective features compared to directly applying convolutions over the raw speech signal. For text processing, we use two branches (a DCNN and a Bidirection RNN followed by a DCNN) in parallel where cross attention is introduced to infer the N-gram level correlations on hidden representations received from the Bi-RNN. Following existing state-of-the-art, we evaluate the performance of the proposed system on the IEMOCAP dataset. Experimental results indicate that the proposed system outperforms existing methods, achieving 3.5% 1 improvement in weighted accuracy.

show abstract

Detection of Human Emotion from Speech—Tools and Techniques

Cited by 3 publications

References 6 publications

Smiling and Non-smiling Emotion Recognition Based on Lower-half Face using Deep-Learning as Convolutional Neural Network

Smiling and Non-smiling Emotion Recognition Based on Lower-half Face using Deep-Learning as Convolutional Neural Network

Representation Learning with Spectro-Temporal-Channel Attention for Speech Emotion Recognition

Attention Driven Fusion for Multi-Modal Emotion Recognition

Contact Info

Product

Resources

About