| 摘要 |
FIELD: physics, acoustics.SUBSTANCE: invention relates to audio signal processing. A device for generating an output signal has at least two output channels and two input channels for an input signal; the device also includes a unit for decomposition into background/forward signals, a background modification unit and a combination unit, wherein the unit for decomposition into background/forward signals is configured to decompose at least two input channels of an input signal, each of which at least decomposes into a signal from a first group of signals and a signal from a second group of signals. The background modification unit is configured to modify a signal from the group of background signals or a signal extracted from a signal from the group of background signals to obtain a modified signal as a first output channel. A combination unit is configured to combine a signal from the group of background signals or a signal extracted from a signal from the group of background signals, and a signal from the group of forward signals or a signal extracted from a signal from the group of forward signals as a second output channel.EFFECT: high accuracy of recognising audio signals.19 cl, 20 dwg |
