Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech
Speaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a well-researched task and there are solutions that provide excellent performance, the...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/3/414 |
| _version_ | 1797446649985892352 |
|---|---|
| author | Nikola Simić Siniša Suzić Tijana Nosek Mia Vujović Zoran Perić Milan Savić Vlado Delić |
| author_facet | Nikola Simić Siniša Suzić Tijana Nosek Mia Vujović Zoran Perić Milan Savić Vlado Delić |
| author_sort | Nikola Simić |
| collection | DOAJ |
| description | Speaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a well-researched task and there are solutions that provide excellent performance, the classification accuracy of developed models significantly decreases when applying them to emotional speech or in the presence of interference. Furthermore, deep models may require a large number of parameters, so constrained solutions are desirable in order to implement them on edge devices in the Internet of Things systems for real-time detection. The aim of this paper is to propose a simple and constrained convolutional neural network for speaker recognition tasks and to examine its robustness for recognition in emotional speech conditions. We examine three quantization methods for developing a constrained network: floating-point eight format, ternary scalar quantization, and binary scalar quantization. The results are demonstrated on the recently recorded SEAC dataset. |
| first_indexed | 2024-03-09T13:43:35Z |
| format | Article |
| id | doaj.art-bfa0ba9560ea4ccb8a310f93b3a49240 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T13:43:35Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-bfa0ba9560ea4ccb8a310f93b3a492402023-11-30T21:03:20ZengMDPI AGEntropy1099-43002022-03-0124341410.3390/e24030414Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional SpeechNikola Simić0Siniša Suzić1Tijana Nosek2Mia Vujović3Zoran Perić4Milan Savić5Vlado Delić6Faculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, SerbiaFaculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, SerbiaFaculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, SerbiaFaculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, SerbiaFaculty of Electronic Engineering, University of Nis, Aleksandra Medvedeva 14, 18000 Nis, SerbiaFaculty of Sciences and Mathematics, University of Pristina in Kosovska Mitrovica, Ive Lole Ribara 29, 38220 Kosovska Mitrovica, SerbiaFaculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, SerbiaSpeaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a well-researched task and there are solutions that provide excellent performance, the classification accuracy of developed models significantly decreases when applying them to emotional speech or in the presence of interference. Furthermore, deep models may require a large number of parameters, so constrained solutions are desirable in order to implement them on edge devices in the Internet of Things systems for real-time detection. The aim of this paper is to propose a simple and constrained convolutional neural network for speaker recognition tasks and to examine its robustness for recognition in emotional speech conditions. We examine three quantization methods for developing a constrained network: floating-point eight format, ternary scalar quantization, and binary scalar quantization. The results are demonstrated on the recently recorded SEAC dataset.https://www.mdpi.com/1099-4300/24/3/414speaker recognitionconvolutional neural networkquantizationemotional speech |
| spellingShingle | Nikola Simić Siniša Suzić Tijana Nosek Mia Vujović Zoran Perić Milan Savić Vlado Delić Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech Entropy speaker recognition convolutional neural network quantization emotional speech |
| title | Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech |
| title_full | Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech |
| title_fullStr | Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech |
| title_full_unstemmed | Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech |
| title_short | Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech |
| title_sort | speaker recognition using constrained convolutional neural networks in emotional speech |
| topic | speaker recognition convolutional neural network quantization emotional speech |
| url | https://www.mdpi.com/1099-4300/24/3/414 |
| work_keys_str_mv | AT nikolasimic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT sinisasuzic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT tijananosek speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT miavujovic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT zoranperic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT milansavic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech AT vladodelic speakerrecognitionusingconstrainedconvolutionalneuralnetworksinemotionalspeech |