Speech Recognition Systems: From Statistical Models to Neural Networks
Keywords:
Automatic Speech Recognition (ASR), Deep Neural Networks (DNN), Neural Language modelling, Hidden Markov Models (HMM), Acoustic-Phonetic ApproachAbstract
The development of the Speech Recognition Systems has been greatly advanced as compared to the initial statistical and rule-based framework to the neural network architecture. This review gives a thorough discussion of the development of the Automatic Speech Recognition (ASR) in terms of acoustic-phonetic and pattern recognition systems, N-gram models, Hidden Markov Models (HMM), and Gaussian Mixture Models (GMM). It points to the drawbacks of statistical analysis, including the scarcity of data, and the lack of contextual modeling, which encouraged people to switch to neural language models. The paper also discusses the development of Deep Neural Networks (DNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Transformer-based architectures as well as end-to-ends models like CTC and attention mechanisms. Other recent developments such as self-supervised learning, hybrid systems and multilingual ASR are also explained using a wide literature review. The paper highlights the achievement in the performance in terms of a reduction in the Word Error Rate (WER) and presents the main challenges, such as noise robustness, low-resource languages, computational efficiency and ethical concerns.
References
[1] C. Yeh, S. Chen, and K. Liao, “Application of Diverse Testing to Improve Integrated Circuit Test Yield and Quality,” Eng, vol. 5, 2024.
[2] I. Ilie and J. Machado, “Design and Validation of a Testing 4D Mechatronic System for Measurement and Integrated Control of Processes,” Machines, vol. 10, 2022.
[3] C. Yeh, J. Chen, C. Chang, and T. Huang, “Using Enhanced Test Systems Based on Digital IC Test Model for the Improvement of Test Yield,” Electronics, vol. 11, 2022.
[4] A. Siddiqui, M. Y. I. Zia, and P. Otero, “A Novel Process to setup Electronic Products Test Sites based on Figure of Merit and Machine Learning,” IEEE Access, 2021, doi: 10.1109/ACCESS.2021.3084545.
[5] N. M. Manousakis, “Advanced Electrical Measurements Technologies,” Energies, vol. 15, 2022.
[6] P. C. Rathebe and M. Kholopo, “Instruments and Measurement Techniques to Assess Extremely Low-Frequency Electromagnetic Fields,” Sensors, vol. 25, 2025.
[7] J. Holub, J. Svatos, and J. Vedral, “Teaching of electronic circuits for measuring technology at the Faculty of electrical engineering, department of measutrement, CTU in Prague,” Meas. Sensors, vol. 38, 2025, doi: 10.1016/j.measen.2024.101312.
[8] M. J. Cordill, O. Glushko, and B. Putz, “electro-Mechanical Testing of conductive Materials Used in Flexible electronics,” Front. Mater., vol. 3, 2016, doi: 10.3389/fmats.2016.00011.
[9] B. K. Prabhu, K. Bhandarkar, D. S. K. N, and D. A. K. M, “Analysis of Functional and Parametric Testing Approaches in Automated Semiconductor Test Systems,” Int. J. Eng. Res. Technol., vol. 14, no. 06, 2025.
[10] L. K. E and N. Palecha, “Pre-Silicon Validation Techniques to Improve Robustness of an IP,” Int. J. Eng. Res. Technol., vol. 11, no. 07, 2022.
[11] M. Gradišnik, T. Beranič, and M. Turkanović, “A Review of Secondary Study on the Verification and Validation of Blockchain Applications: Preliminary Results,” CEUR Work. Proc., 2025.
[12] K. Ma, J. Liu, and M. Molinas, “Guest Editorial: Special Issue on Design and Validation Methodologies for Power Electronics Components and Systems,” IEEE J. Emerg. Sel. Top. POWER Electron., vol. 12, no. 6, 2024, doi: 10.1109/JESTPE.2024.3496972.
[13] J. Ralyté, G. Koutsopoulos, and J. Stirna, “Verification, validation, and evaluation of modeling methods: experiences and recommendations,” Softw. Syst. Model., 2025, doi: 10.1007/s10270-025-01304-2.
[14] S. Khadar, A. M. Kaddouri, A. Kouzou, A. Hafaifa, R. Kennel, and M. Abdelrahem, “Experimental Validation of Different Control Techniques Applied to a Five-Phase Open-End Winding Induction Motor,” Energies, vol. 16, 2023.
[15] M. E. Lonescu and F. A. Stoica, “Verification and Testing Techniques for Reliable System on Chip Solutions,” J. Integr. VLSI, Embed. Comput. Technol., vol. 2, no. 2, 2025, doi: 10.31838/JIVCT/02.02.07.
[16] G. N. Nayak, “ELECTRONICS MANUFACTURING TESTING: A TECHNICAL DEEP DIVE,” Int. J. Comput. Eng. Technol., vol. 16, no. 1, 2025.
[17] R. Entienza, J. Sarmiento, and M. A. T. Mercado, “DEVELOPMENT OF ARDUINO-BASED ELECTRONIC COMPONENT TESTING DEVICE,” Int. J. Adv. Res. Comput. Sci., vol. 15, no. 5, 2024.
[18] P. Garg, B. Gupta, A. Sar, G. Graham, and A. P. Shore, “Development and validation of an instrument to measure the perceived benefits of digitalization in manufacturing,” IEEE Trans. Eng. Manag., 2024.
[19] R. Rhayha and A. A. Ismaili, “Development and validation of an instrument to evaluate the perspective of using the electronic health record in a hospital setting,” BMC Med. Inform. Decis. Mak., vol. 24, 2024.
[20] G. Boboyev and G. Mirpayzieva, “Modern technologies of calibration with measuring devices of electrical quantities,” E3S Web Conf., vol. 461, 2023.
[21] P. Y. Kovalenko, V. I. Mukhin, and M. D. Senyuk, “Development of a methodology for data validation in power systems using different types of measurements,” E3S Web Conf., vol. 288, 2021.
[22] S. P. Pedrozo, V. S. Aquino, V. R. Amorim, R. Q. Peixe, and P. C. Fonseca, “Improving Testing Process to Maximize Validation Efficiency on Advanced Telecommunications Laboratory: An Experience Report,” in The 5th International Conference on Management Engineering, So?ware Engineering and Service Sciences, 2021. doi: 10.1145/3459012.3459014.
