Speech emotion recognition using Dual-Conv2D architecture

• Autorzy: Ayadi Souha , Lachiri Zied

Identyfikatory

DOI

10.15199/48.2024.06.43

Warianty tytułu

PL

Rozpoznawanie emocji mowy przy użyciu architektury Dual-Conv2D

• Języki publikacji: EN

Abstrakty

EN

The ability to convey emotions through speech is still of interest to the field of research. Where different neural network architectures have been developed to be able to automatically recognize emotions. In this work, the main objective is to develop an accurate neural network architecture for Speech emotion recognition. The work includes two main parts which essentially concern the use of MFCC as a feature extractor. And present a new technique for creating a CNN architecture based on the use of two separate architectures based on the Conv2D model.

PL

Możliwość przekazywania emocji za pomocą mowy jest nadal przedmiotem zainteresowania badaczy. Gdzie opracowano różne architektury sieci neuronowych, aby móc automatycznie rozpoznawać emocje. Głównym celem tej pracy jest opracowanie dokładnej architektury sieci neuronowej do rozpoznawania emocji związanych z mową. Praca składa się z dwóch głównych części, które zasadniczo dotyczą wykorzystania MFCC jako ekstraktora cech. Oraz przedstawić nową technikę tworzenia architektury CNN opartą na wykorzystaniu dwóch odrębnych architektur bazujących na modelu Conv2D.

Słowa kluczowe

EN

speech emotion recognition; MFCC; Conv2D

PL

rozpoznawanie emocji mowy; MFCC; Conv2D

• Wydawca: Wydawnictwo SIGMA-NOT
• Czasopismo: Przegląd Elektrotechniczny
• Rocznik: 2024
• Tom: R. 100, nr 6
• Strony: 209--211
• Opis fizyczny: Bibliogr. 13 poz., rys.

Twórcy

autor

Ayadi Souha

• Signal Image and Information Technology(SITI) Laboratory, Department of Electrical Engineering, National Engineering School of Tunis, Campus Universitaire Farhat Hached el Manar BP 37, Le Belvedere 1002 TUNIS

autor

Lachiri Zied

• Signal Image and Information Technology(SITI) Laboratory, Department of Electrical Engineering, National Engineering School of Tunis, Campus Universitaire Farhat Hached el Manar BP 37, Le Belvedere 1002 TUNIS

Bibliografia

• [1] Babak Joze Abbaschian, Daniel Sierra-Sosa, and Adel Elmaghraby. Deep learning techniques for speech emotion recognition, from databases to models. Sensors, 21(4):1249, 2021.
• [2] Tursunov Anvarjon, Mustaqeem, and Soonil Kwon. Deep-net: A lightweight cnn-based speech emotion recognition system using deep frequency features. Sensors, 20(18):5212, 2020.
• [3] UA Asiya and VK Kiran. Speech emotion recognition-a deep learning approach. In 2021 Fifth International Conference on ISMAC (IoT in Social, Mobile, Analytics and Cloud)(I-SMAC), pages 867–871. IEEE, 2021.
• [4] Souha Ayadi and Zied Lachiri. A combined cnn-lstm network for audio emotion recognition using speech and song attributs. In 2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), pages 1–6, 2022.
• [5] Manas Jain, Shruthi Narayan, Pratibha Balaji, Abhijit Bhowmick, Rajesh Kumar Muthu, et al. Speech emotion recognition using support vector machine. arXiv preprint arXiv:2002.07590, 2020.
• [6] Kittisak Jermsittiparsert, Abdurrahman Abdurrahman, Parinya Siriat takul, Ludmila A Sundeeva, Wahidah Hashim, Robbi Rahim, and Andino Maseleno. Pattern recognition and features selection for speech emotion recognition model using deep learning. International Journal of Speech Technology, 23:799– 806, 2020.
• [7] Soonil Kwon et al. 1d-cnn: Speech emotion recognition system using a stacked network with dilated cnn features. Computers, Materials & Continua, 67(3), 2021.
• [8] Manuel Milling, Alice Baird, Katrin D Bartl-Pokorny, Shuo Liu, Alyssa M Alcorn, Jie Shen, Teresa Tavassoli, Eloise Ainger, Elizabeth Pellicano, Maja Pantic, et al. Evaluating the impact of voice activity detection on speech emotion recognition for autistic children. Frontiers in Computer Science, 4:837269, 2022.
• [9] Suprava Patnaik. Speech emotion recognition by using complex mfcc and deep sequential model. Multimedia Tools and Applications, 82(8):11897–11922, 2023.
• [10] Minh H Pham, Farzan M Noori, and Jim Torresen. Emotion recognition using speech data with convolutional neural network. In 2021 IEEE 2nd International Conference on Signal, Control and Communication (SCC), pages 182–187. IEEE, 2021.
• [11] VM Praseetha and PP Joby. Speech emotion recognition using data augmentation. International Journal of Speech Technology, 25(4):783–792, 2022.
• [12] Naris Prombut, Sajjaporn Waijanya, and Nuttachot Promrit. Feature extraction technique based on conv1d and conv2d network for thai speech emotion recognition. In 2021 5th International Conference on Natural Language Processing and Information Retrieval (NLPIR), pages 54–60, 2021.
• [13] Taiba Majid Wani, Teddy Surya Gunawan, Syed Asif Ahmad Qadri, Mira Kartiwi, and Eliathamby Ambikairajah. A comprehensive review of speech emotion recognition systems. IEEE access, 9:47795–47814, 2021.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki i promocja sportu (2025).