FIG. 5 is a figure for describing the reliability of the devices 20 and the final voice recognition result of the voice recognition module 12 of the server apparatus 10. In FIG. 5, results of individually voice-recognizing the speech of Mr. A of “How is the weather today?” recorded with the Internet TV 20A, the Home hub 20B, the desktop PC 20C, and the Laptop PC 20D with the voice recognition module 12 of the server apparatus 10 are “How is the feather Today?”, “How is the weather today?”, “How is the sweater today?”, and “How is the weather today?”, respectively, for example. With respect to a portion where all the voice recognition results are the same in all the devices, for example, the voice recognition module 12 may adopt the portion where the results are the same. With respect to different portions, the weighting based on the evaluation values may be performed (for example, the result in which the evaluation value is equal to or higher than a predetermined value may be adopted or the result in which the evaluation value is the highest may be adopted).

In FIG. 5, the portions of “today” are all the same, and therefore the result may be adopted. The portion of “weather?” includes various words as follows: “feather?”, “weather?”, “sweater?”, and “weather?”, and therefore the results “weather?” of the Home hub 20B (Reliability=13) and the Laptop PC 20D (Reliability=11) with high reliability may be adopted. Then, the voice recognition module 12 may adopt “How is the weather today?” as the final voice recognition result.