SPEECH SIGNAL ENCODING DEVICE, SPEECH SIGNAL DECODING DEVICE, AND SPEECH SIGNAL ENCODING METHOD

摘要

PROBLEM TO BE SOLVED: To obtain speech encoded data having its subjective quality deterioration suppressed even when a speech signal having superposed background noise is inputted. SOLUTION: A voiced/voiceless decision unit 104 decides which of a voiced section that the speech signal is dominant to and a voiceless section that a non-speech signal is dominant to a processed frame of the input speech signal is. When a processed frame which is a voiceless section is inputted as the input speech signal, a filter characteristic of an auditory weight filter 106 is so converted as to suppress a high-frequency band. COPYRIGHT: (C)2003,JPO