A novel speech emotion recognition method based on feature construction and ensemble learning

Ye, Guo; Xiong, Xuejun; Liu, Yangcheng; Li, Xu; Li, Qiong

doi:10.1371/journal.pone.0267132

Cited by 6 publications

(3 citation statements)

References 6 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Authors [25,29] used bagging and boosting ensemble learning approach for stable and improved SER performance. Extreme Gradient Boosting (XGBoost) is popular due to its parallel structure, efficiency, inbuilt capabilities of regularization to avoid overfitting and handling missing values [29,30]. XGBoost aggerates individual models iteratively into ensemble by reducing root mean square error at each step.…”

Section: Experiments and Resultsmentioning

confidence: 99%

Effect of Data Augmentation, Cross-Validation Methods in Robustness of Explainable Speech Based Emotion Recognition

Shinde,

Patil

2024

View full text Add to dashboard Cite

Section: Experiments and Resultsmentioning

confidence: 99%

Effect of Data Augmentation, Cross-Validation Methods in Robustness of Explainable Speech Based Emotion Recognition

Shinde,

Patil

2024

View full text Add to dashboard Cite

“…The accurate and effective extraction of relevant characteristics, as well as the high correlation among these features, are critical elements that significantly affect the effectiveness of the emotion detection system. Contemporary SER approaches have been positively affected by the introduction of several innovative feature extraction methods [ 17 , 18 , 19 , 20 ]. In one study [ 17 ], a deep neural network model for SER that could simultaneously learn both MelSpec and GeMAPS audio features was proposed.…”

Section: Literature Reviewmentioning

confidence: 99%

Enhancing Speech Emotion Recognition Using Dual Feature Extraction Encoders

Pulatov,

Oteniyazov,

Makhmudov

et al. 2023

Sensors

View full text Add to dashboard Cite

Understanding and identifying emotional cues in human speech is a crucial aspect of human–computer communication. The application of computer technology in dissecting and deciphering emotions, along with the extraction of relevant emotional characteristics from speech, forms a significant part of this process. The objective of this study was to architect an innovative framework for speech emotion recognition predicated on spectrograms and semantic feature transcribers, aiming to bolster performance precision by acknowledging the conspicuous inadequacies in extant methodologies and rectifying them. To procure invaluable attributes for speech detection, this investigation leveraged two divergent strategies. Primarily, a wholly convolutional neural network model was engaged to transcribe speech spectrograms. Subsequently, a cutting-edge Mel-frequency cepstral coefficient feature abstraction approach was adopted and integrated with Speech2Vec for semantic feature encoding. These dual forms of attributes underwent individual processing before they were channeled into a long short-term memory network and a comprehensive connected layer for supplementary representation. By doing so, we aimed to bolster the sophistication and efficacy of our speech emotion detection model, thereby enhancing its potential to accurately recognize and interpret emotion from human speech. The proposed mechanism underwent a rigorous evaluation process employing two distinct databases: RAVDESS and EMO-DB. The outcome displayed a predominant performance when juxtaposed with established models, registering an impressive accuracy of 94.8% on the RAVDESS dataset and a commendable 94.0% on the EMO-DB dataset. This superior performance underscores the efficacy of our innovative system in the realm of speech emotion recognition, as it outperforms current frameworks in accuracy metrics.

show abstract

“…Yi Guo, Xuejun Xiong, Yangcheng Liu, Liang Xu, Qiong Li explored the extent of human-computer interaction through ensemble learning methods that are useful for speech accuracy and learner emotion recognition (Guo et al, 2022) (Guo et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Strengthening Music Learning at SMKN

Sulistyowati¹,

Munte²,

Silipta³

et al. 2022

Dimas J. Pemikir. Agama untuk Pemberdaya.

View full text Add to dashboard Cite

The Community Engagement (PkM) of the Faculty of Teacher Training of Christian Education choosing the topic is because: Firstly, the author considers SMKN 1 Pulang Pisau to be a potential high school for activities because their Principal is an arts graduate. Second, the school has complete musical equipment as far as the author is concerned. Third, the location and venue of SMKN 1 Pulang Pisau is spacious and adequate as a location for activities. Fourth, the school was able to command all secondary school principals to send representatives of teachers who had taught in the cultural arts department. We explored creative strategies through an ensemble approach in interfaith music learning after reinforcement and/or empowerment. The implementation method in PkM that we have done is Asset Based Community Development (ABCD) method with implementation stages through discovery, dream, design, define and destiny. Our PkM team evaluated the implementation and sustainability of the program at SMKN 1 Pulang Pisau for 4 days. As a result, the beneficiaries were able to plan, select methods, and implement Cultural Arts learning, with music specifications through an ensemble approach. The ensemble as a national music landscape is able to produce musical creativity and resonant rhythms through materials around the beneficiary schools. For example, gallons, bamboo, and wood.

show abstract

A novel speech emotion recognition method based on feature construction and ensemble learning

Cited by 6 publications

References 6 publications

Effect of Data Augmentation, Cross-Validation Methods in Robustness of Explainable Speech Based Emotion Recognition

Effect of Data Augmentation, Cross-Validation Methods in Robustness of Explainable Speech Based Emotion Recognition

Enhancing Speech Emotion Recognition Using Dual Feature Extraction Encoders

Strengthening Music Learning at SMKN

Contact Info

Product

Resources

About