摘要
提高语音识别系统的鲁棒性是语音识别技术一个重要的研究课题。语音识别系统往往由于训练环境下的数据和识别环境下的数据不匹配造成系统的识别性能下降,为了让语音识别系统在含噪的环境下获得令人满意的工作性能,该文根据人耳听觉特性提出了一种鲁棒语音特征提取方法。在MFCC特征提取之前先对含噪语音特征进行掩蔽特性处理,同时结合语音增强方法对特征进行处理,最后得到鲁棒语音特征。通过4种不同试验结果分析表明,将这种方法用于抗噪声分析可以提高系统的抗噪声能力;同时这种特征的处理方法对不同噪声在不同信噪比有很好的适应性。
Improving the robustness of speech recognition system is an important issue in speech recognition technology. The performance of traditional speech recognition system degrades seriously when the training environments and the testing environments are mismatched. In order to acquire satisfactory performance of speech recognition system under noisy environment, in this essay,a new robust speech feature extraction method based on properties of the human auditory system is presented . This method processes the noisy speech by using masking properties before the MFCC extraction and analyses the speech feature with the speech enhancement algorithm and gets the robust speech feature finally. The results in four different kinds of experiments show that the performance of speech recognition system can be improved greatly by using the new method under noisy environment and the proposed method is highly applicable.
出处
《计算机仿真》
CSCD
2006年第9期80-82,143,共4页
Computer Simulation
关键词
语音识别
噪声
鲁棒性
掩蔽特性
谱减
Speech recognition
Noise
Robustness
Masking model
Spectral substraction
