Abstract:
PURPOSE: A method for removing an interested sound source and a method for recognizing a voice are provided to estimate a mixed noise signal effectively, by removing the interested sound source from a mixed signal provided through two microphones installed in an acoustic signal mixed environment. CONSTITUTION: A voice recognition device initializes a vector(12). The device learns the vector to remove an interested sound source from a mixed signal(10). The device is initialized, and generates a mixed noise signal by removing an interested sound source signal from an input mixed signal(14). When the mixed noise signal is generated, the device generates a mask, by comparing the mixed noise signal with the input mixed signal in a time-frequency domain(16). [Reference numerals] (10) Learn a separation vector(w(K)) for separation of a target sound source; (12) Initialize the separation vector(w(K)); (14) Remove a target sound source; (16) Generate a mask using a mixed noise sound source and an input sound source
Abstract:
PURPOSE: Time delay of signal and a mute sound source dividing method in repercussion environment based on attenuation estimation are provided to consider repercussion environment, estimate per-frequency attenuation and time delay value, and divide mute sound source signal, thereby improve dividing function of the mute sound source signal. CONSTITUTION: Mute sound source divider performs initialization for per-frequency attenuation and time delay value for inputted mixed signal(200). The mute sound source divider learns the per-frequency attenuation and time delay value(202). The mute sound source divider divides per-frequency signal by usage of per-frequency binary mask(206). The mute sound source divider matches order by acquisition of coefficient of correlation of the divided signal(208). The mute sound source divider optimizes the order(210). [Reference numerals] (200) Initializing a per-frequency attenuation and time delay value; (202) Learning a per-frequency attenuation and time delay value; (204) Generating a per-frequency binary mask; (206) Dividing a per-frequency sound source signal by using a per-frequency binary mask; (208) Arranging orders by obtaining correlation coefficient for each per-frequency dividing signal; (210) Order optimization; (AA) Start; (BB) End