DBN-FTLSTM: An Optimized Deep Learning Framework for Speech and Image Recognition
Abstract
Artificial Intelligence's (AI) quick development has brought about a new era of technical innovation with significant ramifications for many different fields. This study explores the development of artificial intelligence (AI) for image and speech recognition, a job in natural language processing that entails the real-time computer transcription of spoken words. The use of deep learning (DL) models for voice recognition has been the subject of several research. But this field is developing quickly. This systematic review offers a thorough and in-depth analysis of research on voice recognition using DL approaches that was published between 2020 and 2024. In order to improve clarity, this study clearly outlines the methodology, including the dataset size, preprocessing methods (MFCC extraction, normalisation, and augmentation), and benchmarking criteria. To provide openness and repeatability, the datasets used for assessment must be specified in reported performance measures (accuracy, RMSE, MAE, and R²). The dataset, experimental setup, and baseline comparisons should all be explicitly stated in order to contextualise the 99.3% accuracy claim. With an overall performance accuracy rate of 97.90%, a root means square error (RMSE) of 0.017, a mean absolute error (MAE) of 0.01, and an R² value of 98.7%, the DBN-FTLSTM model performs better than existing techniques. When compared to other current approaches, the recommended method's error metrics accuracy of 99.3% is the greatest. Mel-frequency cepstral coefficients were the most often used feature extraction approach, while the word error rate was the most commonly used assessment method. Another finding was the dearth of research using transformers, which have been shown to be effective models that can enable parallelization, speed up learning, and enhance the performance of low-resource languages. The findings also identified intriguing and promising areas for further study that had not gotten much attention in previous investigations
DOI:
https://doi.org/10.31449/inf.v49i20.8169Downloads
Published
How to Cite
Issue
Section
License
I assign to Informatica, An International Journal of Computing and Informatics ("Journal") the copyright in the manuscript identified above and any additional material (figures, tables, illustrations, software or other information intended for publication) submitted as part of or as a supplement to the manuscript ("Paper") in all forms and media throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. This transfer includes the right to reproduce and/or to distribute the Paper to other journals or digital libraries in electronic and online forms and systems.
I understand that I retain the rights to use the pre-prints, off-prints, accepted manuscript and published journal Paper for personal use, scholarly purposes and internal institutional use.
In certain cases, I can ask for retaining the publishing rights of the Paper. The Journal can permit or deny the request for publishing rights, to which I fully agree.
I declare that the submitted Paper is original, has been written by the stated authors and has not been published elsewhere nor is currently being considered for publication by any other journal and will not be submitted for such review while under review by this Journal. The Paper contains no material that violates proprietary rights of any other person or entity. I have obtained written permission from copyright owners for any excerpts from copyrighted works that are included and have credited the sources in my article. I have informed the co-author(s) of the terms of this publishing agreement.
Copyright © Slovenian Society Informatika
