Recognition of emotion from speech using evolutionary cepstral coefficients

Bakhshi, Ali; Chalup, Stephan K.; Harimi, Ali; Mirhassani, Seyed Mostafa

doi:10.1007/s11042-020-09591-1

Cited by 6 publications

(5 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…h l i represents the activation value of the i-th neuron in the l-th network. For the backpropagation algorithm, assuming the training sample x (1) , y (1) , x (2) , y (2) , • • • , x (m) , y (m) , the classification label is two classification labels, then the sample cost function is shown in formula 13: J W; x, y = −y log h w (x) − 1 − y log 1 − h w (x) (13) After adding regularization, the overall cost function of the sample set is (Bakhshi et al, 2020):…”

Section: ) Artificial Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Emotion Recognition Algorithm Application Financial Development and Economic Growth Status and Development Trend

Wang

Yan

2022

Front. Psychol.

View full text Add to dashboard Cite

Financial market and economic growth and development trends can be regarded as an extremely complex system, and the in-depth study and prediction of this complex system has always been the focus of attention of economists and other scholars. Emotion recognition algorithm is a pattern recognition technology that integrates a number of emerging science and technology, and has good non-linear system fitting capabilities. However, using emotion recognition algorithm models to analyze and predict financial market and economic growth and development trends can yield more accurate prediction results. This article first gives a detailed introduction to the existing financial development and economic growth status and development trend forecasting problems, and then gives a brief overview of the concept of emotion recognition algorithms. Then, it describes the emotion recognition methods, including statistical emotion recognition methods, mixed emotion recognition methods, and emotion recognition methods based on knowledge technology, and conducts in-depth research on the three algorithm models of statistical emotion recognition methods, they are the support vector machine algorithm model, the artificial neural network algorithm model, and the long and short-term memory network algorithm model. Finally, these three algorithm models are applied to the financial market and economic growth and development trend prediction experiments. Experimental results show that the average absolute error of the three algorithms is below 25, which verifies that the emotion recognition algorithm has good operability and feasibility for the prediction of financial market and economic growth and development trends.

show abstract

Section: ) Artificial Neural Networkmentioning

confidence: 99%

“…After adding regularization, the overall cost function of the sample set is ( Bakhshi et al, 2020 ):…”

Section: Financial Development and Economic Growth Status And Develop...mentioning

confidence: 99%

Emotion Recognition Algorithm Application Financial Development and Economic Growth Status and Development Trend

Wang

Yan

2022

Front. Psychol.

View full text Add to dashboard Cite

show abstract

“…Using approaches like the cosine similarity method, the latent semantic space can be further condensed to locate similar words and documents. Neuropsychological, phrase comprehension reviewer choice and research article recommendation (LSA model) semantic categorization clustering of words (Bakhshi et al, 2020;Bernard et al, 2020;Christy et al, 2020;Gaonkar, 2019). On the bsite2, you may view examples of LSA in action.…”

Section: Latent Semantic Analysis (Lsa)mentioning

confidence: 99%

Deep Learning Speech Synthesis Model for Word/Character-Level Recognition in the Tamil Language

Rajendran

Raja

Nagarajan³

et al. 2023

International Journal of E-Collaboration

View full text Add to dashboard Cite

As electronics and the increasing popularity of social media are widely used, a large amount of text data is created at unprecedented rates. All data created cannot be read by humans, and what they discuss in their sphere of interest may be found. Modeling of themes is a way to identify subjects in a vast number of texts. There has been a lot of study on subject-modeling in English. At the same time, millions of people worldwide speak Tamil; there is no great development in resource-scarce languages such as Tamil being spoken by millions of people worldwide. The consequences of specific deep learning models are usually difficult to interpret for the typical user. They are utilizing various visualization techniques to represent the outcomes of deep learning in a meaningful way. Then, they use metrics like similarity, correlation, perplexity, and coherence to evaluate the deep learning models.

show abstract

“…The Interactive Emotional Dyadic Motion Capture (IEMOCAP) database [37] is a multi-modal emotional corpus. Here, we used four basic emotion labels: anger, happiness, sadness, and neutral in our experiments.…”

Section: Emotional Databasesmentioning

confidence: 99%

“…The Berlin emotional corpus (Emo-DB) [36] and the Enterface05 [35] are freely available for public. The Interactive Emotional Dyadic Motion Capture (IEMOCAP) database [37] is also freely available based on an online request form. These datasets have been referenced in the text as recommended by their publishers.…”

Section: Declarationsmentioning

confidence: 99%

Speech Emotion Recognition using Sub-Band Spectrogram fusion and Deep Convolutional Neural Network transfer learning

Mansouri

Ghaffary

Harimi

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Speech emotion recognition (SER) is a challenging field of research that has attracted research during the last two decades. Successful performance of Deep Convolutional Neural Networks (DNNs) in various difficult pattern recognition problems motivates researchers to develop SER systems based on deep learning algorithms. The most essential requirement in training a deep model is the presence of a large-scale dataset. However, in many cases, such an amount of data is not available. Transfer learning approaches provide a practical solution to this problem. In this paper, we proposed an SER system based on AlexNet, the well-known deep model trained on the large-scale ImageNet dataset. In this way, the novel enriched spectrogram calculated based on the fusion of wide-band and narrow-band spectrograms is developed as a proper input for such a model. The proposed fused spectrogram benefited from both high temporal and spectral resolution. These images have been applied to the pre-trained AlexNet. All the experiments were performed on the popular Emo-DB, IEMOCAP, and eNTERFACE05 datasets based on 10-fold cross-validation and Leave-One-Speaker-Group-Out known as speaker-dependent and speaker-independent techniques, respectively. The proposed approach gains competent performance in contrast to other state-of-the-art methods.

show abstract

Recognition of emotion from speech using evolutionary cepstral coefficients

Cited by 6 publications

References 71 publications

Emotion Recognition Algorithm Application Financial Development and Economic Growth Status and Development Trend

Emotion Recognition Algorithm Application Financial Development and Economic Growth Status and Development Trend

Deep Learning Speech Synthesis Model for Word/Character-Level Recognition in the Tamil Language

Speech Emotion Recognition using Sub-Band Spectrogram fusion and Deep Convolutional Neural Network transfer learning

Contact Info

Product

Resources

About