| 摘要 |
<p>PROBLEM TO BE SOLVED: To obtain a highly precise voice input employing an inexpensive voice inputting section, to understand each of the speeches even though plural riders simultaneously speak without increasing the amount of computations, to recognize a speaker from the device side and to provide precise control. SOLUTION: In a plural rider equipment voice controller 10 installed in plural rider equipment, a speaker estimating section 11 receives voice inputs from voice input sections 1-1 to 1-n to estimate the position of a speaker who makes a voice input and controls the sections 1-1 to 1-n so that the voice in the vicinity of the speaker position is selectively taken in using the estimated speaker position information. A voice understanding section 12 analyzes the voice signals of the speaker obtained through the section 11 so as to understand the contents of the speech. A control section 13 executes an appropriate control based on the voice understanding result of the section 12 and speaker position information.</p> |
