| 摘要 |
PROBLEM TO BE SOLVED: To obtain a desired component from a sound signal including sound from a plurality of sound sources, without impairing sound quality characteristics of the sound signal.SOLUTION: The signal processing device includes: an extracting unit for extracting a component for each sound source that constitutes a frequency spectrum of a sound signal including sound from a plurality of sound sources; a classification unit for classifying each component extracted by the extracting unit, into two or more classes according to a predetermined classification condition; an adjusting unit for adjusting characteristics of an adaptive filter by using a first synthesized spectrum to be synthesized from components classified into a first class by the classification unit, and a second synthesized spectrum to be synthesized from components classified into a second class by the classification unit; and a filtering unit for applying the adaptive filter whose characteristics are adjusted by the adjusting unit to the sound signal. |
