Open AccessSpeech noise Reduction and Automatic Recognition Based on Active Detection Algorithm
Author(s) -
Jianli Mo,
Chenyu Wu
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1607/1/012114
Subject(s) - computer science , speech recognition , voice activity detection , noise (video) , quality (philosophy) , key (lock) , voice analysis , speaker recognition , sound quality , noise reduction , speech processing , artificial intelligence , philosophy , computer security , epistemology , image (mathematics)
In practice, people need to carry out intelligent speech analysis on the sound collected or recorded in some specific scenes. These sounds are recordings of a period of time in some specific places. Due to the influence of the phonetic environment and the limitation of acquisition or recording means, the actual effective and available voice content is very small, and the voice quality is relatively poor. Therefore, it is necessary to enhance the speech of the acquired recordings, eliminate the noise and the interference of non-human voice, and improve the understanding degree of the target’s voice. Voice quality is improved, and the voice can be targeted content recognition and keywords detected, the audio files into a text file, by setting the key words found early warning, the valuable data information in mass speech quick preview, retrieval, analysis, according to the established goals and objectives by using automatic speech recognition in order to improve work efficiency.