论文部分内容阅读
小波变换作为一种新型的信号处理方法，因其可以对信号进行多分辨率分析并有高效算法，所以在DSP上得到了广泛应用。利用语音信号的相关性，可以使用小波变换来压缩语音模板。实践表明，引入小波变换后，在基本保持系统识别率的前提下，节省模板存储空间，从而达到在相同空间下保存更多词条的目的。
As a new type of signal processing method, wavelet transform has been widely used in DSP because of its multi-resolution analysis and efficient algorithm. Using the correlation of speech signals, wavelet transform can be used to compress the speech template. Practice shows that the introduction of wavelet transform, the basic system to maintain the recognition rate under the premise of saving template storage space, so as to achieve the same space to save more entries purposes.