论文部分内容阅读
对高噪声背景下的语音特点进行了分析，提出了适应于高噪声背景下孤立词语音识别的全套方法，进行了系统实现。对本方法和系统进行了详细的实验比较和测试，证明系统在80dB左右接近同类无噪声识别系统的精度，最高可以抵抗100dB的噪声。
The characteristics of speech in the background of high noise are analyzed, and a complete set of methods adapted to the speech recognition of isolated words in the background of high noise is proposed. The system is implemented. The method and system are compared and tested in detail. It is proved that the system can approach the accuracy of similar noiseless identification system around 80dB and can resist the noise up to 100dB.