使用离散小波变换压缩语音模板

Using Discrete Wavelet Transform to Compress Speech Template

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 signal processing method, wavelet transform has been widely used on DSP, for it can analyze signal with multi-resolution and has efficient algorithm. By making use of the correlation of speech signal, discrete wavelet transform can be applied to compress speech template.Practice also shows that the transform can save Flash memory space effectively, but reduces the recognition rate little.