| 摘要 |
<p>PURPOSE:To obtain reproduced sounds for fast listening and slow listening with high quality by dividing the audio signal into frames with fixed length, to search a zero cross-point, and to perform compression and expansion of the frame in accordance with this point. CONSTITUTION:The compression and expansion process of the time base is performed for a digital audio signal stored in a memory 3 by a general purpose signal processing chip 4 called as a DSP. First of all, the audio signal is segmented, then the maximum amplitude of frame is calculated. Next, the zero-cross point is decided after the threshold of an amplitude level difference before and after the zero-cross point is decided. The decision condition is that the product of certain sample data and one previous sample data is zero or minus, and that the difference is pulse values less than the threshold level. Then, a thinning-out process for the fast listening of two times speed and for the compression of 1/2 is performed after a flag for the thinning-out process is set, and the reproduction is made through a filter 6 after an analog conversion.</p> |
