A mobile terminal of the speech recognition method and its device

Technical Field

This application relates to the technical field of information processing, in particular to a mobile terminal based speech recognition method and the corresponding device.

Background Art

The use of the mobile terminal is inseparable from the man-machine interaction process. In intelligent mobile terminal of human-computer interaction more common way is through finger touching the screen of the mobile terminal, by the mobile terminal a built-in inductor touch of the finger to realize the interactive information. With the series of products in the apple Company Siri iPhone added to the voice assistant function, man-machine interactive mode changes from the traditional physical touch for voice control, the instruction language to human meet the demand of the users of the mobile terminal task reached. The speech recognition process allows the user to form with natural language voice assistant software command, relevant device of the mobile terminal after receiving the instruction, by voice assistant software in local and/or clouds server for voice recognition and semantic analysis, and according to the identification and the results of the analysis provide feedback.

However, because the existing speech recognition, in particular semantic analysis technique is not perfect, recognition accuracy is low, especially for multi-words, long sentence, multi-sentence identification and analysis error rate is very high, often the identification and the results of the analysis to differ materially from the user actually needs, the needs of the users input repeatedly, continuously revised recognition and the results of the analysis, seriously affected the mobile terminal-based method for identifying the speech recognition accuracy and rapidity.

Content of the invention

In order to solve the above technical problem, this embodiment provides a mobile terminal of the speech recognition method and corresponding device, based on the mobile terminal in order to improve the speech recognition accuracy and rapidity.

Mobile terminal provided by the application a speech recognition method includes:

Receiving operation of the mobile terminal to be operated of the operation of the triggering message category, the operation class of the mobile terminal based on the business function and of the users of the mobile terminal class division of the range of use; the operation types include: Contact Person category, category of the application program, music categories, search category web page;

Receiving a voice key word information, in the information from the voice key words of the voice keywords is determined, according to the voice key word search category of operation of the timer key thesaurus under, to return to a search result;

The receiving operation of the mobile terminal to be operated in a trigger message categories comprises:

Judging 1st AUDIOMONITOR the Z shaft is monitored whether the weight of the gravity acceleration of 0 to 4 in a range of acceleration of gravity unit, X, Y axis is a gravity acceleration component is in the 4 to 10 in the context of a gravity acceleration units, and the distance of the monitor to the 2nd AUDIOMONITOR whether the balance is zero, the X, Y axis is the plane of the panel of the mobile terminal, the Z axis is perpendicular to X, Y-axis of the plane, the 1st AUDIOMONITOR to the sensor to receive services of the registration sensor of gravity of the monitor, the 2nd AUDIOMONITOR to the sensor to receive the registration server to the distance sensor DETECTAPHONE; if the are is, it is determined that the received operation of the mobile terminal class of trigger message to be operated, the operation class for Contact Person; the receiving voice key word information, in the information from the voice key words of the voice keywords is determined, according to the voice key word search category of operation of the timer key thesaurus under, return to search results include:

Receiving a voice key word information Contact Person, information from the voice keywords in keyword Contact Person determined, according to the library Contact Person Contact Person key word search, the search to return the call to the Contact Person and Contact Person.

Preferably, the receiving operation of the mobile terminal to be operated of the operation of the triggering message categories comprises:

On the screen of the mobile terminal class window rendering operation, when the operation class window corresponding to an operation of a label category click or determined to focus, the operation of the mobile terminal to be operated of the operation of the triggering message categories.

Furthermore, preferably, the operation class window corresponding to the operation of the tag including the types used for realizing communication service function label Contact Person, used for realizing the application of the business function label application program, music playing service function used for realizing a music label and/or is used for realizing the function of on-line search service website search label.

Furthermore, preferably, according to the stated Contact Person to the keyword search the Contact Person includes a plurality of, numbering for each Contact Person, receiving number voice information, voice information corresponding to the calling number Contact Person.

Preferably, when the mobile terminal is operating the vehicle, increase in its operation under operation categories in the key word corresponding to the frequency of the key word, in according to the voice key word search key thesaurus to be operated when the under, in accordance with key word frequency times the order of from large to small the retrieval key thesaurus.

Preferably, when the mobile terminal is after operation, when the the meets preset conditions according to the result of the operation under the type of operation to update the voice key thesaurus.

The application to the mobile terminal of the voice recognition device comprises: a trigger message receiving unit, the voice key word information receiving unit, the voice key word recognition unit and key thesaurus retrieving unit, wherein:

The trigger message receiving unit, for receiving operation of the mobile terminal class of trigger message to be operated, the operation class of the mobile terminal based on the business function and of the users of the mobile terminal class division of the range of use, the operation types include: Contact Person category, category of the application program, music categories, search category web page;

The voice key word information receiving unit, for receiving the voice key word information;

The voice key word recognition unit, from the voice keywords used for determining the number of the voice keywords in the information;

The key thesaurus retrieving unit, according to the voice used for the key word search to the classification operation key thesaurus, to return to a search result;

The trigger message receiving unit includes specifically: monitoring results judging sub-unit and a trigger message receiving sub-unit, wherein:

Said monitoring results judging sub-unit, is used for judging the 1st DETECTAPHONE monitor to the Z axis on the whether the weight of the gravity acceleration of 0 to 4 in a range of acceleration of gravity unit, X, Y axis is a gravity acceleration component is in the 4 to 10 in the context of a gravity acceleration units, and the distance of the monitor to the 2nd AUDIOMONITOR whether the balance is zero, the X, Y axis is the plane of the panel of the mobile terminal, the Z axis is perpendicular to X, Y-axis of the plane, the 1st AUDIOMONITOR to the sensor to receive services of the registration sensor of gravity of the monitor, the 2nd AUDIOMONITOR to the sensor to receive the registration server, the distance sensor of the monitor;

The trigger message receiving sub-unit, is used for the judging result is, the operation of the mobile terminal class of trigger message to be operated, the operation class for Contact Person;

The voice key word information receiving unit for receiving a voice key word information Contact Person, the voice key word recognition unit from the voice used in particular for determining the key word information Contact Person keyword, the keyword searching unit in particular for on the basis of the key word search Contact Person Contact Person library, to return to the retrieved Contact Person;

The device also includes the call unit, for call the retrieved Contact Person.

Preferably, the trigger message receiving unit includes specifically: operation categories window presents the subunit and trigger message receiving sub-unit, wherein:

The operation class window presentation sub-unit, used in the rendering operation of the mobile terminal class window on the screen;

The trigger message receiving sub-unit, the operation class for the one operation of the window corresponding to the category click or a label is determined to be the focus, the mobile terminal received the operation class of trigger message to be operated.

Furthermore, preferably, the device also includes Contact Person number unit and number voice information receiving unit, wherein: said Contact Person number unit, used for the key word Contact Person according to the retrieved when Contact Person includes a plurality of, numbering for each Contact Person; the number of voice information receiving unit, for receiving a number of voice information, the call unit in particular for call number corresponding to the voice information Contact Person.

Preferably, the device also comprises a key word frequency increase in unit, the mobile terminal is used in the operation, increase in its operation under operation categories in the key word corresponding to the frequency of the key word, the key word for the specific search unit, according to the voice key word search key thesaurus to be operated when the under, in accordance with key word frequency times the order of from large to small the retrieval key thesaurus.

Preferably, the device also comprises a key word update unit, the mobile terminal is used in the operation, when the the meets preset conditions according to the operation of the results of the classification operation to update the key thesaurus.

The application according to an embodiment of the services of the mobile terminal is received a functional classification class of operation after the trigger message, receiving a voice key word information, key word from the pronunciation of the voice keywords is determined, according to the voice key word is then retrieves the corresponding key thesaurus, and return to the search results. With the existing speech recognition art, the application according to the embodiment of the business function of the dividing operation, with only the key thesaurus corresponding to each operation categories, on the one hand, at the time of search according to the voice key word searching processing object is limited to the operation of the mobile terminal with the corresponding key thesaurus, reduce the number of the processing object, the processing capacity of the mobile terminal a is weak; and on the one hand, the retrieval to reduce the number of the processing objects the shortening of time for the retrieval process, thereby improving the efficiency of the speech recognition; in a further aspect, a processing object retrieval to the reduction in the number of the repeated and key word ambiguity probability is reduced, thereby improving the accuracy of the speech recognition. Moreover, in this embodiment the receiving voice information in the form of voice key word information received, no longer is the ordinary natural language, avoiding multi-words, long sentence and multi-sentence, more easily, on the one hand from the speech information extract new keyword, and the efficiency of the speech recognition is improved; on the other hand, by a key word from the pronunciation of the key word extracting information key thesaurus matching to obtain the return result, help to improve the accuracy of speech recognition.

Description of drawings

In order to more clearly illustrate the application or embodiment of the technological scheme in the prior art, the solid will be

Construct in the description or the prior art to be used for the simple introduction of the Figure, it is obvious that,

The Figure is described below in this application is only recorded in some embodiments, common in the art

In terms of technical personnel, on the premise of a creative work, can also be based on these with photos to

The Figure shall be other.

Figure 1 is the flowchart of the mobile terminal of this application a speech recognition one embodiment of the method;

Figure 2 is the structure diagram of the application of speech recognition of the mobile terminal one embodiment of the device.

Mode of execution

In order to ensure that the technical personnel in the area of geographical xie Ben better technical proposal in the application, the application will be in connection with the embodiment of the Figure, in the present embodiment a clear the technical scheme, complete described, obviously, the embodiment described is only a portion of the embodiment of this application, , but not all, of the embodiment. Based on the embodiment of the in this application, one of ordinary skill in the art without making creative work is obtained on the premise that all of the other embodiments, an application should be within the scope of protection.

See Figure 1, the illustration of the application of the mobile terminal of the embodiment of the voice identification method for the flow. The process comprises:

Step S101: the operation of the mobile terminal receives treats holds of the operation of the triggering message category, the operation class of the mobile terminal based on the categories of service function;

With the development of information technology, not only the mobile terminal having the traditional communication function alone, but also has many new business function, for example, Internet search, play audio-video, such as playing games. These different business function of the differences of the nature of the, mobile terminal users to achieve various business function mode of operation, the operation instruction. In spite of this, to achieve the same operational function with various operating normally, according to the embodiment of different service function of the mobile terminal in advance of the operation of the various possible category. Through this kind of operation types of dividing the follow-up of the targeted with a clear speech recognition process. This embodiment does not limit the operation of the identified category number and type, as long as they can meet the actual application needs. For example, according to the mobile terminal itself can be the business function and of the users of the mobile terminal divided by the range of use of the following categories: category Contact Person, is used for storing the names of Contact Person, telephone number, information such as the personal characteristics, in speech recognition, the user can view a certain Contact Person the relevant information to the Contact Person, can call the Contact Person, such as sending messages to the Contact Person; application program categories, name of the application program used for recording, icon, such as the storage position information associated with the application, one of the speech recognition of the application program when the application program can be viewed the basic attribute of the information, the application program can carry out various operations: starting, unloading, delete, update, and the like; the music category, used for recording music name, Singer name, relevant information such as the album name, in the voice recognition a certain music on the music can be viewed when the basic attribute information, the music can be various operating: playing, mobile, delete; website search categories, search function is used for realizing the webpage.

Step S102: receiving a voice key word information, in the information from the voice key words of the voice keywords is determined;

If required the user of the mobile terminal to the mobile terminal using voice realize some of the control, operation, the speech recognition engine can be started, so that it is in a working state, when the needs of the speech recognition, the speech recognition engine receives the voice key word information. This embodiment receiving voice information is a keyword to the theme of the voice content, may not be intended to general includes the complete sentence of the natural language. For example, if one needs to be certain phone, the voice is of the prior art: "phone to XX", in the case of this embodiment, when it is determined that the category information of operation when "Contact Person", it is able to directly tell "a certain", need only provide the keyword operation, it can control the operation of the corresponding mobile terminal.

Received after the voice key word information, from the voice key word information to determine in the voice key word. Voice information of the users of the mobile terminal cannot normally very accurately only voice keywords, for example, may include some transition sound, capable of tone, these voice for speech recognition purposes which belongs to the noise, from the voice key word information need be removed in, voice keywords are extracted, the voice keywords in directly corresponding to the key thesaurus a certain keyword, then corresponds to a certain operation command.

Step S103: according to the voice key word search operation of the timer under the operation of the key thesaurus categories, to return to a search result;

Through the above steps after determining the voice keywords, the using the key word to be operated corresponding to the operation of the key thesaurus category in the search, and return to the search results. After the acquisition of the retrieval results, the retrieval results can be triggered the implementation of the corresponding operation of the mobile terminal.

It should be explained that: in this embodiment the steps S101 and S102 in the actual operating process can be performed in parallel in the operating or S102 before the step S101 after the step, the user of the mobile terminal can be triggered as noted previously the operation of the first category to be operated, and then receives user-input voice key word; receiving can also be voice keywords of the user, the receiving user with the operation of the operation class of the trigger, or in receiving treatment of operation types of trigger the voice key word information also receives, this execution timing between the two does not affect the aim of the invention, according to the application needs, wherein an appropriate mode can be selected.

This embodiment received the services of the mobile terminal based on the functional classification of a certain operation categories after the triggering message, receiving a voice key word information, key word from the pronunciation of the voice keywords is determined, according to the voice key word is then retrieves the corresponding key thesaurus, and return to the search results. With the existing speech recognition art, the embodiment of the following technical effects can be achieved:

(1) because the categories according to the business function to carry out the division operation, with only the key thesaurus corresponding to each operation categories, this is different from existing speech recognition for use with various different operating properties, all of the speech recognition library way, according to the voice keywords to the search processing of the target at the time of search will be limited to only the mobile terminal to carry out corresponding to the operation range of the key word, reduce the number of the processing object, the processing capacity of the mobile terminal a is weak. For example, existing speech recognition library comprising 100 speech operation instruction, to the the embodiment 100 of a voice operation instructions of the type, used for realizing the order of the function "Contact Person" attributable to a category, the category includes 10 a voice operation instructions, when the mobile terminal user when the function of only needs Contact Person, it will trigger in the category of speech search identification, that is, only needs to be in this 10 a search instruction, the voice operation, therefore, greatly reduce the number of processing.

(2) because of search relates to a decrease in the number of a processing object, the processing capacity of a mobile terminal under the condition of not changing, the accomplishment of a search will be greatly shorten the time of the process, within a relatively short period of time with the user's input can be provided after the voice key word corresponding to the search result, thereby improving the efficiency of the speech recognition. To explain to the precedent, hypothesis searching each voice operation instruction time is 0.01s, the user speaking a pronunciation word of the position is located in the subsection 80 bit, in accordance with the existing voice recognition mode, in the above-mentioned 100 speech operation in instruction base 80 after matching search to find the voice operation instructions, for time 0.8s, however, if the search matching operation in achieving the function of restrictions Contact Person 10 a within the range of voice operation instructions, is however only the largest 0.1s, we can see that the retrieval time is greatly shortened, thereby improving the efficiency of the speech recognition.

(3) because of search relates to a processing object is the reduction in the number of the repeated and key word ambiguity probability is reduced, thereby improving the accuracy of the speech recognition. For example, the user speaking the word "sheet of certain", in the above-mentioned 100 in a voice operation instructions, possible to find two "sheet of certain", a "sheet of certain" by a user on the mobile terminal is stored on the name of a Contact Person, a "sheet of certain" user's music library is stored in the name of a Singer, that is to say, the speech word exists and repeats and ambiguous, then the system will not know whether the user of the mobile terminal in the telephone directory to the "sheet of certain" of a telephone call, or to listen to the music library of songs "a certain", if default choose the former, then user actually the idea of realizing the latter may be; if the default selection of the latter, the user really is to realize the idea of the possible. However, in this embodiment, the user types the operation is specified in advance, if the designated category for "Contact Person", the user said "sheet of certain", that is, to a certain telephone with the sheet; if the designated category is "music", the user said "sheet of certain", that is, the want to listen to the songs of a certain card, so as to accurately carry out the speech recognition operation.

(4) in this embodiment the receiving voice information in voice received in the form of a key word information, no longer is the ordinary natural language, avoiding multi-words, long sentence and multi-sentence, more easily, on the one hand from the speech information extract new keyword, and the efficiency of the speech recognition is improved; on the other hand, by a key word from the pronunciation of the key word extracting information key thesaurus matching to obtain the return result, help to improve the accuracy of speech recognition.

Mentioned in the aforesaid embodiment of the mobile terminal needs to be received to be operated of the operation of the triggering message category, in the process of practical application, receives the trigger message in several ways. For example, when the user needs to use the speech recognition engine operation control mobile terminal, the mobile terminal screen that appears on one of the categories window, window in that category in the category label displaying various operation, the class label can include: used for realizing communication service function label Contact Person, used for realizing the application of the business function label application program, music playing service function used for realizing a music label, used for realizing the function of on-line search service webpage searching, and so on. When a user clicks on a label in these categories, or the focal point moves to the label a certain class, will be produced in the system a trigger event (trigger message), monitoring the triggering event that can be received by the triggering message to the operating types. Also for example, when the user is provided with the application program of the automatic updating function, when the found a certain appears in the network to the new version of the application program, the mobile terminal receiving the update notification, that is received at this time can be considered to the update notice "application program" this operation the trigger message of the categories, thus can receive user's voice instruction to realize the application program update or are not updated. Furthermore, in addition to the above based on a touch-control events or network event to as receiving to the operating types of trigger information, the user can also be based on certain habitual movements of the mobile terminal to determine whether to receive the trigger message of operation categories. A common action, such as a user of the mobile phone is placed to the ear, the action that the user need to call a certain Contact Person, in this case, it may be that the received "Contact Person" category. This kind of trigger mode specific process is as follows:

In the speech recognition engine at the time of initialization to service the sensor of the system, the registering a gravity sensor monitor and a distance sensor of the monitor, the gravity sensor can be provided in three dimensions of the gravity acceleration (x, y, z) of the component. When the mobile phone is placed horizontally, along a gravity acceleration value z shaft tend to 9.8, and x, the weight of the shaft tends to y 0. Therefore, voice assistant application real-time monitoring gravity acceleration sensor return value, when the mobile phone is placed horizontally or slightly inclined time (that is, user's normal grasps evenly the mobile phone) z tend to the weight of the shaft 7, and at the same time to judge the distance sensor is a non-return value of 0 (that is, the mobile phone of the distance sensor from the front without any object), meets the above 2 the whole process is an initialization conditions, and recording initialization time. Get one's ear of the user of the mobile phone before the distance sensor in the process always to return to non -0 (without any obstacles) value, the state is working. When the user of the mobile phone when placing one's ear, z shaft at this time tend to 2 (need to prove, the value may be in the 0 to 4 the gravitational acceleration the unit can be full of the invention purpose of the full application), y x shaft and of the shaft and of the absolute value of the tend to 7 (the value can be in 4 to 10 in the value range), taking into account the user the mobile phone is placed ear x shaft is provided with a tilted angle, x should be at this time is larger than the absolute value of the shaft 2 the, the system meets the above conditions and working state, interacting WAIT_PROXI the state of the system, this state waiting for the distance sensor to return to 0 value (face block the distance sensor), once returned to 0 value will start-up procedures Contact Person dialing operation of the call, if the distance sensor to return to 0 value before, the whole process from the initialized to WAIT_PROXI more than 2 seconds, the judgment of the failure identifying actions. When the call after starting Contact Person dialing function, the user may directly call the name of Contact Person, system from a mobile phone according to the recognition result list Contact Person Contact Person meet the conditions is read in, if there are a plurality of matching Contact Person, through the voice prompt the user to the system, for example (1. Certain Chen. 2. mr./Ms. Liu ), the user only need to say that "the 1 or [...]" 2 the dial [...] can select certain or Wang XX to Chen, when the user selects a rear, the system will prompt the user to dial the ongoing, and to the direct dial Contact Person selected by a user. If only one Contact Person, the system will directly indicate the user is to carry out dialing and dial telephone.

In the above-mentioned embodiment is not limited to the voice key word in the category specific how to realize under the operation of the retrieval key thesaurus, although this does not affect the aim of the invention. However, with a user in long-term use in the process of a voice recognition function, a corollary with the regularity of customary, these habits can be applied to the retrieval process to the key thesaurus. For example, when the mobile terminal often when a certain operation is carried out, the need for this kind of operation of the user demand relatively frequent, at this time, a counter can be set, the mobile terminal is recorded is carried out a certain operation of the operation is carried out after the total number of times (frequency), the total number of in as a key word with the action corresponding to an attribute of the keyword, searching on the basis of the voice keywords, in accordance with the size of the frequency of the keyword to the sequence of the retrieval key thesaurus, since the user often carry out a certain operation, the operation inevitably larger the frequency, in the front key thesaurus, the retrieval from large to small will be relatively rapid in order to obtain the retrieval result. Furthermore, the mobile terminal can also be after operation, the preset condition is satisfied according to the result of the operation under the type of operation to update the voice key thesaurus. For example, for the increase in the list Contact Person a person, then the need to update the voice key thesaurus, will the increased Contact Person as key words in the key thesaurus, more new time can be increased every time at the time of after one Contact Person, can also be each time when the restarting of the mobile phone, these can be set according to the actual situation, when the pre-set condition is satisfied, that is, trigger the update operation.

Detailed description of the above-mentioned contents of the mobile terminal of this application embodiment of a method of speech recognition, correspondingly, the application also provides a mobile terminal device embodiment of speech recognition. See Figure 2, the illustration of the mobile terminal of this application of speech recognition block diagram the structure of the device. The device comprises: a trigger message receiving unit 201, voice key word information receiving unit 202, voice key word recognition unit 203 and key thesaurus retrieving unit 204, wherein:

Trigger message receiving unit 201, is used for receiving operation of the mobile terminal class of trigger message to be operated, the operation class of the mobile terminal based on the categories of service function;

Voice key word information receiving unit 202, is used for receiving the voice key word information;

Voice key word recognition unit 203, from the voice keywords used for determining the number of the voice keywords in the information;

Key thesaurus retrieving unit 204, is used to search the voice keywords to the classification operation key thesaurus, to return to a search result.

The above-mentioned device is the working process of the embodiment of: the trigger message receiving unit 201 receives the operation of the mobile terminal class of trigger message to be operated; the voice key word information receiving unit 202 receives voice key word information, a voice key word recognition unit 203 in the voice key word information from voice keywords is determined; then, by the key word searching unit 204 according to the voice key word search category of operation of the timer key thesaurus under, to return to a search result.

The device according to the embodiment of the services of the mobile terminal is received a functional classification class of operation after the trigger message, receiving a voice key word information, key word from the pronunciation of the voice keywords is determined, according to the voice key word is then retrieves the corresponding key thesaurus, and return to the search results. With the existing speech recognition art, the device according to the embodiment of the business function of the dividing operation, with only the key thesaurus corresponding to each operation categories, on the one hand, at the time of search according to the voice key word searching processing object is limited to the operation of the mobile terminal with the corresponding key thesaurus, reduce the number of the processing object, the processing capacity of the mobile terminal a is weak; and on the one hand, the retrieval to reduce the number of the processing objects the shortening of time for the retrieval process, thereby improving the efficiency of the speech recognition; in a further aspect, a processing object retrieval to the reduction in the number of the repeated and key word ambiguity probability is reduced, thereby improving the accuracy of the speech recognition. Moreover, the embodiment of the device to the receiving voice information in the form of voice key word information received, no longer is the ordinary natural language, avoiding multi-words, long sentence and multi-sentence, more easily, on the one hand from the speech information extract new keyword, and the efficiency of the speech recognition is improved; on the other hand, by a key word from the pronunciation of the key word extracting information key thesaurus matching to obtain the return result, help to improve the accuracy of speech recognition.

In the process of practical application, is provided with a plurality of triggering operation types of mode, different ways of the trigger message receiving unit corresponding to the specific structure may be different. Two ways are provided below, the technicians of this field based on the two way other implementations can be derived:

One of the ways: by pop-window and receives the user to click or in the movement of the focal point method to determine the received triggering message operation categories. This way, the trigger message receiving unit 201 can include: window operation categories appear subunit 2011 and a trigger message receiving sub-unit 2012, wherein:

Operation categories window presenting subunit 2011, used in the rendering operation of the mobile terminal class window on the screen;

Trigger message receiving sub-unit 2012, the operation class for the one operation of the window corresponding to the category click or a label is determined to be the focus, the mobile terminal received the operation class of trigger message to be operated.

Mode II: the identification of the user through the inductor such that the mode of operation to the operation class triggering message is received. This way, the trigger message receiving unit includes specifically: monitoring results judging sub-unit and a trigger message receiving sub-unit, wherein:

Said monitoring results judging sub-unit, is used for judging the monitor to the 1st AUDIOMONITOR Z axis acceleration of gravity on whether the weight is 2, X, Y axis is acceleration of gravity whether the weight is 7, and the distance of the monitor to the 2nd AUDIOMONITOR whether the balance is zero, the X, Y axis is the plane of the panel of the mobile terminal, the Z axis is perpendicular to X, Y-axis of the plane, the 1st AUDIOMONITOR to the sensor to receive services of the registration sensor of gravity of the monitor, the 2nd AUDIOMONITOR to the sensor to receive the registration server, the distance sensor of the monitor;

The trigger message receiving sub-unit, is used for the judging result is, the operation of the mobile terminal class of trigger message to be operated, the operation class for Contact Person.

In 2nd way, other function of the change of the unit there is a corresponding, i.e., speech specific key word information receiving unit for receiving a voice key word information Contact Person, voice keywords used for specific recognition unit from the voice key word information keyword Contact Person determined in, key word searching unit in particular for on the basis of the key word search Contact Person Contact Person library, to return to the retrieved Contact Person. Embodiments also include the above-mentioned device the call unit, for call the retrieved Contact Person. Furthermore, the above-mentioned device embodiments also include Contact Person number unit and number voice information receiving unit, wherein: said Contact Person number unit, used for the key word Contact Person according to the retrieved when Contact Person includes a plurality of, numbering for each Contact Person; the number of voice information receiving unit, for receiving a number of voice information, the call unit in particular for call number corresponding to the voice information Contact Person.

Furthermore, can also be based on some practical needs, to the above-mentioned an embodiment of a device to replace certain deformation or equivalent, in order to obtain the technical effect of a more optimized. For example, the above-mentioned an embodiment of a device also comprises a key word frequency increase in unit, the mobile terminal is used in the operation, increase in its operation under operation categories in the key word corresponding to the frequency of the key word, the key word for the specific search unit, according to the voice key word search key thesaurus to be operated when the under, in accordance with key word frequency times the order of from large to small the retrieval key thesaurus. The unit can be improved by increasing the speed of the search. As another example, the above-mentioned an embodiment of a device may also include a key word update unit 205, the mobile terminal is used in the operation, when the the meets preset conditions according to the result of the operation under the type of operation to update the key word.

It should be explained that: in order to describe the simple and convenient, the specification of the above-mentioned embodiment and various deformation of the embodiment of the focus in the realization mode with all of the other embodiment or deformation mode is different, the same similar between the various cases refer to each other can be the part of the. In particular, the device of the embodiment of the case of several improved manner, similar to the embodiment of the method, is relatively simple so described, refer to the related method embodiment can be part of the description. The device described in the above embodiment of the unit can be or may not be physically separate, can be located in a local, or also can be distributed to a plurality of network environment. In the process of practical application, can be selected according to the actual needs of the part or the whole of the unit to realize the purpose of the embodiment, one of ordinary skill in the art in the case of paying creative work, can understand and implement.

The stated above is the concrete mode of execution of this application, it should be noted that, in the technical field as the ordinary technical personnel, without deviating from the principle of this application, can also be made a number of improvements and retouches, these improvements and retouches should also be regarded as the scope of protection of this application.