In the server apparatus 10, the voice recognition module 12 and the voice assistant module 13 can be configured by modules of hardware or software (programs, such as OS, API, service, utility, library, driver, and application) or a combination thereof, for example. The computer may realize the functions of the voice recognition module 12 and the voice assistant module 13 by executing the programs.

As illustrated in FIG. 1, the plurality of devices 20 each has a communication portion 21 (or communication module 21) for performing data communication with the server apparatus 10 through the network 30, a microphone 22 for inputting a voice, and a request portion 23 (or request module 23) recording a user's speech through the microphone 22, and then transmitting data of the recorded user's speech (recorded data) to the server apparatus 10 through the communication module 21, and an output unit 24.

The output unit 24 contains one or two or more of a monitor, such as a liquid crystal display, an organic EL display, a plasma display, and an CRT, a headset containing a speaker and an earphone or a headphone, a printer, and the like and has a function of outputting information and a voice to the outside. The type and the specification of the mounted output unit 24 vary depending on the devices 20.

The microphone 22 includes a single microphone, a dual microphone, array microphones, and the like, for example, and collects and inputs voices, such as a user's speech.