Research on the Introduction to Models used in Speech Recognition

DOI

10.2991/assehr.k.220504.363How to use a DOI?

Keywords

Artificial Intelligence; Speech Recognition; HMM; ANN

Abstract

The theoretical basis of Artificial Intelligence (AI) has been established since the middle of the last century. After 1970, after John McCarthy and others established an AI laboratory at MIT, the first batch of scholars and technologies began to emerge, and AI began to flourish. path of. Today, AI is still the focus and frontier of today’s science and technology. After decades of technological precipitation, AI covers a wider range of aspects, and the knowledge system is huge and increasingly complex. Therefore, this article will introduce the basic principles and models of AI from the perspective of speech recognition. By consulting the relevant literature, this article will specifically introduce the hidden Markov model and the artificial neural network model. After the summary of these papers, we found that the hidden Markov model has very good time series modeling ability, but the classification ability is not strong. The latter has excellent classification capabilities, but cannot describe time-varying information well. Therefore, this article combines the two models. After consulting the literature, I finally found that the combination of the two models is a good choice and has great potential for development.

Copyright

© 2022 The Authors. Published by Atlantis Press SARL.

Open Access

This is an open access article distributed under the CC BY-NC 4.0 license.

Download article (PDF)

TY  - CONF
AU  - Xinsheng Wu
PY  - 2022
DA  - 2022/06/01
TI  - Research on the Introduction to Models used in Speech Recognition
BT  - Proceedings of the 2022 8th International Conference on Humanities and Social Science Research (ICHSSR 2022)
PB  - Atlantis Press
SP  - 2014
EP  - 2020
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.220504.363
DO  - 10.2991/assehr.k.220504.363
ID  - Wu2022
ER  -