Subhayu Ghosh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Subhayu Ghosh has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Nanda Dulal Jana
Department of Computer Science and Engineering, National Institute of Technology Durgapur, Durgapur, India
8 shared publications Nanda Dulal Jana
Department of Computer Science and Engineering, National Institute of Technology, Durgapur, India
3 shared publicationsSandipan Dhar
National Institute of Technology Durgapur
2 shared publications- Anubhav Dwivedi
Department of Computer Science and Engineering, National Institute of Technology Durgapur, Durgapur, India
1 shared publication - Arjun Ghosh
Department of Computer Science and Engineering, National Institute of Technology Durgapur, Durgapur, India
1 shared publication Brahmanand Dubey
National Institute of Technology Durgapur
1 shared publication
Latest Publications
- 22 February 2026
- Published by Springer Nature
- p. 333-345
ViT-GA-AVSS: Optimizing Vision Transformer-Based Audio-Visual Speech Synthesis Approach Using Genetic Algorithm
- 22 February 2026
- Published by Springer Nature
- p. 333-345
ViT-GA-AVSS: Optimizing Vision Transformer-Based Audio-Visual Speech Synthesis Approach Using Genetic Algorithm
- Top Cited Papers
- 16 January 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 309
Intelligent neural architecture search via Taguchi design and language model-based differential evolution for agricultural image recognition
- 16 January 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 309
Top Cited PapersIntelligent neural architecture search via Taguchi design and language model-based differential evolution for agricultural image recognition
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1966-1975
Enhanced Audio-Visual Speech Synthesis via Multi-Discriminative Learning
Audio-visual speech synthesis (AVSS) aims to produce an audio-visual stream that conveys a target speaker's speech. In this study, the AVSS system takes the input speech of a source speaker and generates the audio-visual stream of the target speaker while preserving the linguistic content of the source speech. The process involves two main components: voice conversion (VC), which adapts the vocal features from the source to the target speaker, and audio-visual synthesis (AVS), which generates the synchronized audio-visual stream from the transformed speech. This paper presents a novel generative framework based on multi-discriminative learning to enhance the realism and quality of AVSS outputs. The proposed approach integrates multiple discriminators, including capsule networks, co-occurrence neural networks, and vision transformers (ViTs), within the VC model to leverage their unique strengths in capturing diverse speech features. Additionally, the AVS model incorporates a co-occurrence neural network to improve video quality and achieve better temporal alignment between audio and visual data. Experimental evaluations on standard benchmarks demonstrate that the proposed method achieves significant improvements in both audio and video quality, offering a substantial advancement in AVSS technology.
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1966-1975
Enhanced Audio-Visual Speech Synthesis via Multi-Discriminative Learning
Audio-visual speech synthesis (AVSS) aims to produce an audio-visual stream that conveys a target speaker's speech. In this study, the AVSS system takes the input speech of a source speaker and generates the audio-visual stream of the target speaker while preserving the linguistic content of the source speech. The process involves two main components: voice conversion (VC), which adapts the vocal features from the source to the target speaker, and audio-visual synthesis (AVS), which generates the synchronized audio-visual stream from the transformed speech. This paper presents a novel generative framework based on multi-discriminative learning to enhance the realism and quality of AVSS outputs. The proposed approach integrates multiple discriminators, including capsule networks, co-occurrence neural networks, and vision transformers (ViTs), within the VC model to leverage their unique strengths in capturing diverse speech features. Additionally, the AVS model incorporates a co-occurrence neural network to improve video quality and achieve better temporal alignment between audio and visual data. Experimental evaluations on standard benchmarks demonstrate that the proposed method achieves significant improvements in both audio and video quality, offering a substantial advancement in AVSS technology.
- 14 October 2025
- Published by Springer Nature in Journal of Heuristics
- Vol. 31 (4) , 1-51
Automated CNN architecture design with enhanced particle swarm optimization
- 14 October 2025
- Published by Springer Nature in Journal of Heuristics
- Vol. 31 (4) , 1-51
Automated CNN architecture design with enhanced particle swarm optimization
- 8 January 2025
- Published by Springer Nature in Journal of Molecular Modeling
- Vol. 31 (2) , 1-11
TransConv: convolution-infused transformer for protein secondary structure prediction
Context Protein secondary structure prediction is essential for understanding protein function and characteristics and can also facilitate drug discovery. Traditional methods for experimentally determining protein structures are both time-consuming and costly. Computational biology offers a viable alternative by predicting protein structures from their sequences. Protein secondary structure is defined by the local spatial arrangement of the protein backbone, resulting from hydrogen bonds between amino acids. Methods In this study, we introduce TransConv, a model that combines transformer models with convolutional blocks to predict protein secondary structures from amino acid sequences. Transformer models are effective at capturing long-range dependencies through self-attention mechanisms. We integrate convolutional blocks into the transformer architecture to improve the detection of important local features. This hybrid model captures both long-range interactions and local features, leading to more accurate predictions of protein secondary structures, thus offering an efficient solution for this critical task. The experimental outcomes on the benchmark datasets depict the superiority of the proposed approach over the state-of-the-art (SOTA) models in the literature.
- 8 January 2025
- Published by Springer Nature in Journal of Molecular Modeling
- Vol. 31 (2) , 1-11
TransConv: convolution-infused transformer for protein secondary structure prediction
Context Protein secondary structure prediction is essential for understanding protein function and characteristics and can also facilitate drug discovery. Traditional methods for experimentally determining protein structures are both time-consuming and costly. Computational biology offers a viable alternative by predicting protein structures from their sequences. Protein secondary structure is defined by the local spatial arrangement of the protein backbone, resulting from hydrogen bonds between amino acids. Methods In this study, we introduce TransConv, a model that combines transformer models with convolutional blocks to predict protein secondary structures from amino acid sequences. Transformer models are effective at capturing long-range dependencies through self-attention mechanisms. We integrate convolutional blocks into the transformer architecture to improve the detection of important local features. This hybrid model captures both long-range interactions and local features, leading to more accurate predictions of protein secondary structures, thus offering an efficient solution for this critical task. The experimental outcomes on the benchmark datasets depict the superiority of the proposed approach over the state-of-the-art (SOTA) models in the literature.
- Open Access
- 18 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 10-18
Audio-Visual Speech Synthesis Leveraging Capsule-Enhanced Generative Adversarial Network
- 18 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 10-18
Open AccessAudio-Visual Speech Synthesis Leveraging Capsule-Enhanced Generative Adversarial Network
- 18 November 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (4) , 1-18
Active Learning with Particle Swarm Optimization for Enhanced Skin Cancer Classification Utilizing Deep CNN Models
Skin cancer is a critical global health issue, with millions of non-melanoma and melanoma cases diagnosed annually. Early detection is essential to improving patient outcomes, yet traditional deep learning models for skin cancer classification are often limited by the need for large, annotated datasets and extensive computational resources. The aim of this study is to address these limitations by proposing an efficient skin cancer classification framework that integrates active learning (AL) with particle swarm optimization (PSO). The AL framework selectively identifies the most informative unlabeled instances for expert annotation, minimizing labeling costs while optimizing classifier performance. PSO, a nature-inspired metaheuristic algorithm, enhances the selection process within AL, ensuring the most relevant data points are chosen. This method was applied to train multiple Convolutional Neural Network (CNN) models on the HAM10000 skin lesion dataset. Experimental results demonstrate that the proposed AL-PSO approach significantly improves classification accuracy, with the Least Confidence strategy achieving approximately 89.4% accuracy while using only 40% of the labeled training data. This represents a substantial improvement over traditional approaches in terms of both accuracy and efficiency. The findings indicate that the integration of AL and PSO can accelerate the adoption of AI in clinical settings for skin cancer detection. The code for this study is publicly available at (https://github.com/Sayantani-31/AL-PSO).
- 18 November 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (4) , 1-18
Active Learning with Particle Swarm Optimization for Enhanced Skin Cancer Classification Utilizing Deep CNN Models
Skin cancer is a critical global health issue, with millions of non-melanoma and melanoma cases diagnosed annually. Early detection is essential to improving patient outcomes, yet traditional deep learning models for skin cancer classification are often limited by the need for large, annotated datasets and extensive computational resources. The aim of this study is to address these limitations by proposing an efficient skin cancer classification framework that integrates active learning (AL) with particle swarm optimization (PSO). The AL framework selectively identifies the most informative unlabeled instances for expert annotation, minimizing labeling costs while optimizing classifier performance. PSO, a nature-inspired metaheuristic algorithm, enhances the selection process within AL, ensuring the most relevant data points are chosen. This method was applied to train multiple Convolutional Neural Network (CNN) models on the HAM10000 skin lesion dataset. Experimental results demonstrate that the proposed AL-PSO approach significantly improves classification accuracy, with the Least Confidence strategy achieving approximately 89.4% accuracy while using only 40% of the labeled training data. This represents a substantial improvement over traditional approaches in terms of both accuracy and efficiency. The findings indicate that the integration of AL and PSO can accelerate the adoption of AI in clinical settings for skin cancer detection. The code for this study is publicly available at (https://github.com/Sayantani-31/AL-PSO).
- 8 August 2024
- Published by Springer Nature
- p. 303-318
A Comprehensive Analysis on Features and Performance Evaluation Metrics in Audio-Visual Voice Conversion
- 8 August 2024
- Published by Springer Nature
- p. 303-318
A Comprehensive Analysis on Features and Performance Evaluation Metrics in Audio-Visual Voice Conversion
- 8 August 2024
- Published by Springer Nature
- p. 384-396
An Ensemble of Machine Learning Models Utilizing Deep Convolutional Features for Medical Image Classification
- 8 August 2024
- Published by Springer Nature
- p. 384-396
An Ensemble of Machine Learning Models Utilizing Deep Convolutional Features for Medical Image Classification