Voice processing apparatus, voice processing system, and voice processing program

A voice processing system provided in a plurality of locations for processing voice signals generated by a plurality of microphones in a situation in which a voice emitted from a first sound source is collected by the plurality of microphones and in which a voice emitted from a second sound source including the voice emitted by the first sound source and collected as an echo component is collected by the plurality of microphones, the system comprising:

• a band dividing portion dividing at least a first voice signal and a second voice signal generated by at least a first microphone and a second microphone among the plurality of microphones, respectively, into predetermined frequency bands;
• a sound source segregating portion segregating the echo component of the voice emitted by the first sound source included in the voice emitted by the second sound source in each of the predetermined frequency bands of the first and second voice signals thus divided by the band dividing portion based on the power of the voices collected by the first and second microphones;
• and
• a band synthesis portion synthesizing the first and second voice signals from which the echo component of the first sound source has been segregated by the sound source segregating portion into a voice signal including the voice emitted by the first sound source and a voice signal including the segregated echo component of the first sound source.

A voice processing program for processing voice signals generated by a plurality of microphones in a situation in which a voice emitted by a first sound source is collected by the plurality of microphones and in which a voice emitted by a second sound source including the voice emitted by the first sound source and collected as an echo component is collected by the plurality of microphones, the program comprising:

• a band dividing process dividing at least a first voice signal and a second voice signal generated by at least a first microphone and a second microphone among the plurality of microphones, respectively, into predetermined frequency bands;
• a sound source segregating process segregating the echo component of the voice emitted by the first sound source included in the voice emitted by the second sound source in each of the predetermined frequency bands of the first and second voice signals divided by the band dividing process based on the power of the voices collected by the first and second microphones;
• and
• a band synthesis process synthesizing the first and second voice signals from which the echo component of the first sound source has been segregated by the sound source segregating process into a voice signal including the voice emitted by the first sound source and a voice signal including the segregated echo component of the first sound source.