Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification
Speaker identification (SI) involves recognizing a speaker from a group of unknown speakers, while speaker verification (SV) determines if a given voice sample belongs to a particular person. The main drawbacks of SI are session variability, noise in the background, and insufficient information. To...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IIUM Press, International Islamic University Malaysia
2025-01-01
|
| Series: | International Islamic University Malaysia Engineering Journal |
| Subjects: | |
| Online Access: | https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/3386 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549913705414656 |
|---|---|
| author | Pravin Marotrao Ghate Bhagvat D. Jadhav Shriram Sadashiv Kulkarni Pravin Balaso Chopade Prabhakar N. Kota |
| author_facet | Pravin Marotrao Ghate Bhagvat D. Jadhav Shriram Sadashiv Kulkarni Pravin Balaso Chopade Prabhakar N. Kota |
| author_sort | Pravin Marotrao Ghate |
| collection | DOAJ |
| description |
Speaker identification (SI) involves recognizing a speaker from a group of unknown speakers, while speaker verification (SV) determines if a given voice sample belongs to a particular person. The main drawbacks of SI are session variability, noise in the background, and insufficient information. To mitigate the limitations mentioned above, this research proposes Global Local Self-Attention (GLSA) based Long Short-Term Memory (LSTM) with Exponential Neighborhood – Grey Wolf Optimization (EN-GWO) method for effective speaker identification using TIMIT and VoxCeleb 1 datasets. The GLSA is incorporated in LSTM, which focuses on the required data, and the hyperparameters are tuned using the EN-GWO, which enhances speaker identification performance. The GLSA-LSTM with EN-GWO method acquires an accuracy of 99.36% on the TIMIT dataset, and an accuracy of 93.45% on the VoxCeleb 1 datasets, while compared to SincNet and Generative Adversarial Network (SincGAN) and Hybrid Neural Network – Support Vector Machine (NN-SVM).
ABSTRAK: Pengenalpastian pembicara (Speaker Identification, SI) melibatkan pengenalan pembicara daripada kumpulan pembicara yang tidak dikenali, manakala pengesahan pembicara (Speaker Verification, SV) menentukan sama ada sampel suara tertentu milik seseorang individu. Kekurangan utama dalam SI ialah variasi sesi, bunyi latar belakang, dan maklumat yang tidak mencukupi. Untuk mengatasi kekangan tersebut, kajian ini mencadangkan kaedah Global Local Self-Attention (GLSA) berasaskan Long Short-Term Memory (LSTM) dengan Pengoptimuman Grey Wolf Jiranan Eksponen (EN-GWO) bagi pengenalpastian pembicara yang berkesan menggunakan set data TIMIT dan VoxCeleb 1. GLSA digabungkan dalam LSTM yang memberi tumpuan pada data yang diperlukan, manakala parameter hiper ditala menggunakan EN-GWO untuk meningkatkan prestasi pengenalpastian pembicara. Kaedah GLSA-LSTM dengan EN-GWO mencapai ketepatan 99.36% pada dataset TIMIT dan ketepatan 93.45% pada dataset VoxCeleb 1, berbanding dengan SincNet dan Generative Adversarial Network (SincGAN) serta Hybrid Neural Network – Support Vector Machine (NN-SVM).
|
| format | Article |
| id | doaj-art-33e2878cbc8c46f5a673c45952bc70ad |
| institution | Kabale University |
| issn | 1511-788X 2289-7860 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IIUM Press, International Islamic University Malaysia |
| record_format | Article |
| series | International Islamic University Malaysia Engineering Journal |
| spelling | doaj-art-33e2878cbc8c46f5a673c45952bc70ad2025-01-10T12:40:39ZengIIUM Press, International Islamic University MalaysiaInternational Islamic University Malaysia Engineering Journal1511-788X2289-78602025-01-0126110.31436/iiumej.v26i1.3386Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker IdentificationPravin Marotrao Ghate0Bhagvat D. Jadhav1https://orcid.org/0000-0002-1393-6823Shriram Sadashiv Kulkarni2https://orcid.org/0000-0001-8584-8171Pravin Balaso Chopade3Prabhakar N. Kota4https://orcid.org/0000-0002-7537-8433JSPM's Rajarshi Shahu College of EngineeringJSPM's Rajarshi Shahu College of EngineeringSinhgad Academy of EngineeringM.E.S College of EngineeringM.E.S College of Engineering Speaker identification (SI) involves recognizing a speaker from a group of unknown speakers, while speaker verification (SV) determines if a given voice sample belongs to a particular person. The main drawbacks of SI are session variability, noise in the background, and insufficient information. To mitigate the limitations mentioned above, this research proposes Global Local Self-Attention (GLSA) based Long Short-Term Memory (LSTM) with Exponential Neighborhood – Grey Wolf Optimization (EN-GWO) method for effective speaker identification using TIMIT and VoxCeleb 1 datasets. The GLSA is incorporated in LSTM, which focuses on the required data, and the hyperparameters are tuned using the EN-GWO, which enhances speaker identification performance. The GLSA-LSTM with EN-GWO method acquires an accuracy of 99.36% on the TIMIT dataset, and an accuracy of 93.45% on the VoxCeleb 1 datasets, while compared to SincNet and Generative Adversarial Network (SincGAN) and Hybrid Neural Network – Support Vector Machine (NN-SVM). ABSTRAK: Pengenalpastian pembicara (Speaker Identification, SI) melibatkan pengenalan pembicara daripada kumpulan pembicara yang tidak dikenali, manakala pengesahan pembicara (Speaker Verification, SV) menentukan sama ada sampel suara tertentu milik seseorang individu. Kekurangan utama dalam SI ialah variasi sesi, bunyi latar belakang, dan maklumat yang tidak mencukupi. Untuk mengatasi kekangan tersebut, kajian ini mencadangkan kaedah Global Local Self-Attention (GLSA) berasaskan Long Short-Term Memory (LSTM) dengan Pengoptimuman Grey Wolf Jiranan Eksponen (EN-GWO) bagi pengenalpastian pembicara yang berkesan menggunakan set data TIMIT dan VoxCeleb 1. GLSA digabungkan dalam LSTM yang memberi tumpuan pada data yang diperlukan, manakala parameter hiper ditala menggunakan EN-GWO untuk meningkatkan prestasi pengenalpastian pembicara. Kaedah GLSA-LSTM dengan EN-GWO mencapai ketepatan 99.36% pada dataset TIMIT dan ketepatan 93.45% pada dataset VoxCeleb 1, berbanding dengan SincNet dan Generative Adversarial Network (SincGAN) serta Hybrid Neural Network – Support Vector Machine (NN-SVM). https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/3386Exponential Neighborhood – Grey Wolf Optimizationlobal-Local Self-AttentionLong Short-Term MemoryMel Frequency Cepstral CoefficientSpeaker Identification |
| spellingShingle | Pravin Marotrao Ghate Bhagvat D. Jadhav Shriram Sadashiv Kulkarni Pravin Balaso Chopade Prabhakar N. Kota Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification International Islamic University Malaysia Engineering Journal Exponential Neighborhood – Grey Wolf Optimization lobal-Local Self-Attention Long Short-Term Memory Mel Frequency Cepstral Coefficient Speaker Identification |
| title | Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification |
| title_full | Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification |
| title_fullStr | Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification |
| title_full_unstemmed | Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification |
| title_short | Global-Local Self-Attention-Based Long Short-Term Memory with Optimization Algorithm for Speaker Identification |
| title_sort | global local self attention based long short term memory with optimization algorithm for speaker identification |
| topic | Exponential Neighborhood – Grey Wolf Optimization lobal-Local Self-Attention Long Short-Term Memory Mel Frequency Cepstral Coefficient Speaker Identification |
| url | https://journals.iium.edu.my/ejournal/index.php/iiumej/article/view/3386 |
| work_keys_str_mv | AT pravinmarotraoghate globallocalselfattentionbasedlongshorttermmemorywithoptimizationalgorithmforspeakeridentification AT bhagvatdjadhav globallocalselfattentionbasedlongshorttermmemorywithoptimizationalgorithmforspeakeridentification AT shriramsadashivkulkarni globallocalselfattentionbasedlongshorttermmemorywithoptimizationalgorithmforspeakeridentification AT pravinbalasochopade globallocalselfattentionbasedlongshorttermmemorywithoptimizationalgorithmforspeakeridentification AT prabhakarnkota globallocalselfattentionbasedlongshorttermmemorywithoptimizationalgorithmforspeakeridentification |