| 摘要 |
PROBLEM TO BE SOLVED: To improve the accuracy of speech recognition in a noisy environment and realize increased speed of speech recognition in a quite environment.SOLUTION: A terminal according to the present invention comprises: a speech acquisition unit for acquiring first speech information; a first sound collection processing unit for removing the noise included in the first speech information using a first removal scheme, and outputting the noise-removed speech information as second speech information; a first speech recognition unit for performing speech recognition on the second speech information, and outputting the result of speech recognition as first speech recognition result information; a communication unit for receiving, from a server as second speech recognition result information, a speech recognition result that is the result of speech recognition performed on noise-removed third speech information, with the noise included in the first speech information removed using a second removal scheme that removes a greater amount of noise from the first speech information than the amount of noise removed from the first speech information by the first removal scheme; and an arbitration unit for selecting which of the first speech recognition result information and the second speech recognition result information will be output.SELECTED DRAWING: Figure 2 |
