FULL-VOICE INTERACTION SYSTEM FOR MOBILE INTELLIGENT DEVICES Field of the Invention The invention relates to a full-voice interaction system for mobile intelligent devices. Background of the Invention Voice recognition technology has been widely used in real life. The google speech recognizing open sdk , the IFlytek speech service, the Baidu voice service and the like can all provide a search service and other services based on voice commands to users. However, these techniques do not get rid of text forms and need visual participation, which is inconvenient on some occasions. Summary of the Invention It is a preferred aim of the invention to provide a full-voice interaction system for mobile intelligent devices, which can eliminate dependence on human vision. In one broad form, the invention provides a full-voice interaction system for mobile intelligent devices, typically based on a mobile terminal, comprising: a voice receiving unit used for receiving external voice information; a voice recognition unit used for recognizing the voice information and forming text formatting information; a semantic analysis unit used for executing semantic analysis on the text formatting information; a command converting unit used for converting the semanteme analyzed by the semantic analysis unit into an executable control instruction; an execution unit used for executing a corresponding action according to the control instruction; a voice synthesis unit used for synthesizing text formatting result information generated by the execution unit into voice format information; and a voice output unit used for outputting the voice format information output by the voice synthesis unit. 1 According to a preferred embodiment of the invention, the text formatting information of the execution unit is synthesized into the voice format information, and the voice format information is played by the voice output unit, namely, people orally command when in use, and a result based on the command can be played in a voice manner, so that full-voice interaction is achieved. Brief Description of the Drawings Fig. 1 is a connection diagram of a preferred embodiment of the invention. Detailed Description of the Embodiments In order that the purposes, technical solutions and advantages of the invention are clearer, a further detailed description of one or more embodiments of the invention will be given below in combination with the accompanying drawings. The invention enables existing product resources on mobile intelligent or 'smart' devices, such as voice reception, voice recognition, semantic analysis, voice synthesis, voice output and the like, to be effectively integrated, in order to operate and start the complicated mobile intelligent devices in an approximately natural language interaction mode, which can be accepted by most people. The invention aims at providing full-voice interaction in order to satisfy user demands, and the invention can achieve the beneficial effects by utilizing the existing products or modules in a novel manner. The invention does not have to rely on software innovation, but can utilize and integrate the existing software instead. For example, the voice receiving unit and the voice output unit can be, respectively, a microphone and a loudspeaker on a mobile terminal. Companies such as Google, Baidu have already transformed voice into texts with an accuracy of higher than 90%, and a free application interface can be used to serve as the voice recognition unit of the invention. Similarly, in voice synthesis technology such as Interphonic, a free application interface 2 has been provided and can be used to serve as the voice synthesis unit. Google and Baidu, for example, provide a free open semantic analysis interface for common users on mobile intelligent devices, which can serve as the semantic analysis unit of the invention. The command converting unit can similarly use common technology of the mobile terminal, for example, existing widely used search engine technology. The execution unit can be any component capable of executing commands in the mobile terminal. Fig. 1 shows a full-voice interaction system for mobile intelligent devices. The mobile intelligent devices can be smartphones, tablet computers and the like, and the system includes: a voice receiving unit 1, used for receiving external voice information, for example, a mobile microphone; a voice recognition unit 2, used for recognizing the voice information received by the voice receiving unit 1 and forming text formatting information; a semantic analysis unit 3, used for executing semantic analysis on the text formatting information output from the voice recognition unit 2; a command converting unit 4, used for converting the semanteme analyzed by the semantic analysis unit 3 into an executable control instruction; an execution unit 5, used for executing a corresponding action according to the control instruction output from the command converting unit 4; the action specifically includes hardware functions of the mobile terminal, such as opening navigation, browsing network news, checking weather and traffic information, opening WIFI and Bluetooth, adjusting volume, and the like, and the above is not exhaustive; a voice synthesis unit 6, used for synthesizing text formatting result information generated by the execution unit 5 into voice fonnat information; and a voice output unit 7, used for outputting the voice fonnat information output by the voice synthesis unit 6, for example a mobile phone loudspeaker. 3 According to an embodiment of the invention, the text formatting information of the execution unit 4 is synthesized into the voice format information, and the voice format information is played by the voice output unit 7, and even if people orally command when in use, a result based on the command can be played in a voice manner in the invention, so that full-voice interaction is achieved. The invention can specifically include, but is not limited to the following purposes: 1. providing convenience for people, who are inconvenient to operate and start the mobile intelligent devices during driving, hiking and the like, to operate electronic devices in a full-voice mode so as to ensure the safety; 2. providing operation convenience for wearable devices, or other mobile intelligent devices with small screens and which are inconvenient to be operated by pressing keys or touching screens; 3. providing convenience for people with poor vision or unfamiliar with the technology to operate and start complicated mobile intelligent devices in an answer interaction mode close to natural language. The foregoing descriptions are embodiments of the invention, rather than limiting the scope of protection of the innovation patent. Any obvious modifications, equivalent substitutions and improvements and the like made within the spirit and principle of the invention shall fall within the scope of protection of the innovation patent. 4