| 摘要 |
<P>PROBLEM TO BE SOLVED: To suppress strangeness in listening. <P>SOLUTION: A sound correction device according to one embodiment includes input means, calculation means, division means, conversion means, extraction means, synthesis means, and generation means. The input means inputs a sound signal propagated in a reproduction acoustic field. The calculation means calculates an impulse response from the sound signal. The division means divides the impulse response into a first time domain including direct sound or initial reflection sound and a second time domain including a reverberant sound component. The conversion means converts each of the divided impulse response into a frequency spectrum. The extraction means specifies a frequency component converted from the second time domain whose amplitude component has a relatively higher peak in comparison with an amplitude component of the frequency spectrum converted from the first time domain, and extracts the frequency component as a resonant component. The synthesis means synthesizes a predetermined characteristic and a characteristic for attenuating the extracted resonant component. The generation means corrects the frequency spectra to the synthesized characteristic. <P>COPYRIGHT: (C)2012,JPO&INPIT |
