INFORMATION PROCESSOR, AND SOUND SOURCE SEPARATION METHOD AND PROGRAM OF INFORMATION PROCESSOR

• Main Authors: ROJVIBOONCHAI KULTIDA, KOIKE ATSUSHI, TAKIZAWA KAZUYUKI
• Format: Patent
• Language: English
• Published: 23.04.2009
• Subjects: ACOUSTICS; DEAF-AID SETS; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKEACOUSTIC ELECTROMECHANICAL TRANSDUCERS; MUSICAL INSTRUMENTS; PHYSICS; PUBLIC ADDRESS SYSTEMS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION; TELEPHONIC COMMUNICATION

PROBLEM TO BE SOLVED: To prevent the degradation of noise removing performance even in switching between a microphone input signal and an EC reference signal. SOLUTION: When an EC function utilization state is changed, a performance degradation preventing section 212 sets the EC function utilization state of operation mode information to "EC switching being in effect" and sets a data change channel to an unused channel. When a terminal utilization state is changed, the performance degradation preventing section 212 sets the terminal utilization state to "shape switching being in effect" and sets the data change channel to an unused channel. Then, after a lapse of T milliseconds from the change of the utilization state, the performance degradation preventing section 212 sets the EC function utilization state and the terminal utilization state to the states after changing and setting the unused channel to "none". During the period till the lapse of T milliseconds, a sound source separating section 211 does not use the unused channel, and changeover between switches SW1 and SW2 occurs after the lapse of T milliseconds. Thus, any incorrect EC processing or incorrect sound source separation processing does not occur, thereby preventing the degradation of the noise removing performance. COPYRIGHT: (C)2009,JPO&INPIT