论文部分内容阅读
一.引言语音识别是一种让机器通过识别和理解过程把语音信号转变为文本或命令的高级技术,涉及到生理学、心理学、语言学、计算机科学以及信号处理等诸多领域。近年来语音识别在视频领域出现了很多应用,如音字转写、固定音频检索、语种识别、音频特征提取、关键词检索等等。应用自动语音识别技术,将大大提高效率并大幅降低成本。语音识别作为一门交叉学科,经过多年的积累研究,获得了巨大的进展。特别是近20年来,语音识别技术取得了显着的进步,并
I. INTRODUCTION Speech recognition is an advanced technique that enables machines to translate speech signals into texts or commands by recognizing and understanding processes involving a variety of fields such as physiology, psychology, linguistics, computer science, and signal processing. In recent years, speech recognition has appeared many applications in the field of video, such as transliteration, fixed audio retrieval, language recognition, audio feature extraction, keyword search and so on. The application of automatic speech recognition technology, will greatly improve efficiency and significantly reduce costs. Speech recognition as a cross-discipline, after years of accumulated research, has made tremendous progress. Especially in the past 20 years, speech recognition technology has made significant progress