Speech processing apparatus and method

1. A speech processing apparatus comprising:

an acquisition processor configured to acquire a speech, wherein a microphone array including a plurality of microphones to acquire the speech and to provide the acquired speech to the acquisition processor;

a separation processor configured to separate the speech into a plurality of sections in accordance with a prescribed rule;

a calculation processor configured to calculate a degree of similarity in each combination of the sections;

an estimation processor configured to estimate, with respect to the each section, a direction of arrival of the speech;

a correction processor configured to group the sections whose directions of arrival are mutually similar into a same group and correct the degree of similarity with respect to the combination of the sections in the same group; and

a clustering processor configured to cluster the sections by using the corrected degree of similarity.

2. The apparatus according to claim 1 , wherein the calculation processor includes:

a feature calculation processor configured to calculate acoustic features of each section, and

a similarity calculation processor configured to calculate the degree of similarity in each combination of the sections by using the calculated acoustic features.

3. The apparatus according to claim 2 ,

wherein the correction processor corrects the calculated degree of similarity with respect to the combination of the sections in the same group to a higher degree when the calculated degree of similarity is higher than a prescribed threshold.

4. The apparatus according to claim 3 ,

wherein the correction processor corrects the calculated degree of similarity to a higher degree by multiplying the calculated degree of similarity by N(N is a real number whose value is more than 1), or by raising the calculated degree of similarity to an M-th power(M is a real number whose value is more than 1).

5. A speech processing method comprising:

acquiring a speech, wherein the acquiring the speech acquires the speech through a microphone array;

separating the speech into a plurality of sections in accordance with a prescribed rule;

calculating a degree of similarity in each combination of the sections;

estimating, with respect to the each section, a direction of arrival of the speech;

grouping the sections whose directions of arrival are mutually similar into a same group and correcting the degree of similarity with respect to the combination of the sections in the same group; and

clustering the sections by using the corrected degree of similarity.

6. A non-transitory computer readable medium including computer executable instructions, wherein the instructions, when executed by a processor, cause the processor to perform a method comprising:

acquiring a speech, wherein the acquiring the speech acquires the speech through a microphone array;

separating the speech into a plurality of sections in accordance with a prescribed rule;

calculating a degree of similarity in each combination of the sections;

estimating, with respect to the each section, a direction of arrival of the speech;

grouping the sections whose directions of arrival are mutually similar into a same group and correcting the degree of similarity with respect to the combination of the sections in the same group; and

clustering the sections by using the corrected degree of similarity.