| 摘要 |
PROBLEM TO BE SOLVED: To realize voice level control in which delay and distortion are suppressed. SOLUTION: Since delay is small when the high pass component d, i.e. the output from an HPF1, out of the frequency components of a voice signal a is large, level control is performed at a mutiplication calculating section 7 and a multiplying section 8 with a fixed multiplier according to the peak value in a zero-cross section similarly to prior art. On contrary, when the low pass component c, i.e. the output from an LPF2, is large, level control is performed at the mutiplication calculating section 7 and the multiplying section 8 by dynamic real time processing so that no delay is generated thus realizing voice level control in which delay and distortion are suppressed. COPYRIGHT: (C)2006,JPO&NCIPI |
