Speech Recognition Technology Methods based on Deep Learning

DOI

10.2991/978-94-6463-823-3_11How to use a DOI?

Keywords

Deep Learning; Speech Recognition; Acoustic Model

Abstract

This paper aims to systematically sort out and construct a theoretical framework in the domain of speech recognition technology. First, this article outlines the current research status of speech recognition technology and its importance on the discipline of artificial intelligence(AI). This article introduces the speech recognition methods based on deep learning in detail, divides them into two categories: acoustic model (AM) and front-end processing (SFE), and further subdivides the specific methods under each type, such as DNN structure, LSTM model and CNN model. This article explains the technical principles, application scenarios and core ideas of each method, and summarizes its derivative methods. In particular, in-depth discussions were conducted on the DNN structure in the acoustic model, the improvement of time series modeling capabilities combined with LSTM-RNN, and the use of attention mechanisms to enhance the concentration effect on crucial speech segments. Additionally, this article also discusses the time-frequency mask estimation in front-end processing and the acoustic robustness enhancement technology based on it. Finally, this paper classifies and prospects various methods and proposes possible future development trends.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jiangyu Luo
PY  - 2025
DA  - 2025/08/31
TI  - Speech Recognition Technology Methods based on Deep Learning
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 114
EP  - 124
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_11
DO  - 10.2991/978-94-6463-823-3_11
ID  - Luo2025
ER  -