A method of signal extraction from noisy signal based on auditory scene analysis

The paper presents a method of extracting the desired signal from a noisy signal by using physical constraints as a model of acoustic source segregation. Using physical constraints, which are related to the four regularities proposed by Bregman, the proposed method can solve the problem of segregating two acoustic sources. The physical constraints are obtained from the regularities by translating from qualitative to quantitative conditions. Three simulations were carried out using the following signals: (a) a noise-added AM complex tone, (b) a mixed AM complex tones, and (c) noisy synthesized vowel. The performance of the proposed method was evaluated using two measures: precision, that is, likely SNR, and spectrum distortion (SD). We found that using signals (a) and (b), it could extract the desired AM complex tone from a noise-added AM complex tone or mixed AM complex tones, in which signal and noise exist in the same frequency region. In particular, the average of the reduced SD was about 20 dB. Moreover, using signal (c), it could also extract the desired speech signal from noisy speech.リサーチレポート（北陸先端科学技術大学院大学情報科学研究科