论文部分内容阅读
语音识别在实际应用中受到信道噪声和便携终端计算、存储能力不足等因素制约。分布式语音识别(DSR)不仅解决了上述的问题,还有占用带宽窄、综合成本低等优点,但其应用的前提是提取的参数必须标准化。文中介绍了分布式语音识别前端处理的基本结构以及相关标准。
In practical applications, speech recognition is restricted by channel noise, calculation of portable terminal, insufficient storage capacity and other factors. Distributed Speech Recognition (DSR) not only solves the above problems, but also has the advantages of narrow occupied bandwidth and low overall cost. However, the application of DSR requires that the extracted parameters must be standardized. This paper introduces the basic structure of distributed speech recognition front-end processing and related standards.