KR20200053341A - Vehicle and controlling method thereof - Google Patents

Abstract

A vehicle according to one disclosed embodiment includes a communication unit communicating with a terminal providing a call function; A voice input unit that receives a user's voice; A storage unit that stores a user pattern related to the user's call pattern; And when a user's voice is input, search for at least one name candidate corresponding to the input voice, and determine a threshold for a confidence score of the at least one name candidate based on a user pattern. , A controller for selecting a name corresponding to the input voice among at least one name candidate based on the determined threshold.

Description

Vehicle and its control method {VEHICLE AND CONTROLLING METHOD THEREOF}

It relates to a vehicle that recognizes a user's voice and performs a specific function according to the recognized voice and a control method thereof.

As the technology of a vehicle develops, various functions for user convenience are provided in addition to driving, which is a basic function performed by the vehicle.

As the functions that the vehicle can perform are diversified, the user's operation load is increased, and the increase in the operation load is a factor that degrades concentration on driving and hinders safe driving.

Accordingly, research and development of a user interface for reducing a user's operation load is required. In particular, when a voice recognition technology that recognizes a user's voice and performs a function corresponding to the voice is applied to the vehicle, the user's It is expected that the operation load can be effectively reduced.

One aspect provides a vehicle that recognizes a user's voice and provides a function corresponding to the user's voice and a control method thereof.

As a technical means for achieving the above-described technical problem, a vehicle according to an aspect includes: a communication unit communicating with a terminal providing a call function; A voice input unit that receives a user's voice; A storage unit that stores a user pattern related to the user's call pattern; And when the user's voice is input, search for at least one name candidate corresponding to the input voice, and set a threshold for a confidence score of the at least one name candidate based on the user pattern. And a controller which selects a name corresponding to the input voice among the at least one name candidate based on the determined threshold value.

In addition, when the phonebook information and the call history information of the terminal are received, the controller may generate a user pattern based on the received phonebook information and call history information.

In addition, the controller may select a name corresponding to the input voice among name candidates whose reliability score exceeds the first threshold.

In addition, when the number of name candidates whose reliability score exceeds the first threshold is 1, the controller determines whether the reliability score of the name candidate exceeds the second threshold, and the reliability of the name candidate If the score exceeds the second threshold, the name candidate can be selected as a name corresponding to the input voice.

In addition, if the reliability score of the name candidate whose reliability score exceeds the first threshold is less than or equal to the second threshold, the controller may request the user to confirm the name candidate.

In addition, when the number of name candidates whose reliability score exceeds the first threshold is plural, the priority of the plurality of name candidates is determined based on the reliability score, and the name candidate whose priority is the first priority And a third score and a difference in the reliability score between the second and second name candidates, and selecting a name corresponding to the input voice based on the comparison result.

In addition, when the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate exceeds the third threshold, the control unit selects the first-ranked name candidate. It can be selected as a name corresponding to the input voice.

In addition, if the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is equal to or less than the third threshold, the user is prompted to confirm the first-ranked name candidate. You can ask.

In addition, when the phonebook information of the terminal is received, the controller searches for at least one name candidate corresponding to the input voice among the received phonebook information, and a first threshold value based on the number of searched name candidates. , At least one of the second threshold value or the third threshold value may be changed.

Also, when the number of searched candidates is less than or equal to a predetermined first reference number, the control unit may increase the second threshold value and decrease the third threshold value.

In addition, the control unit may check the number of categories for the searched name candidate from the phonebook information, and change at least one of a first threshold value, a second threshold value, or a third threshold value based on the number of categories. have.

In addition, when the number of the categories is equal to or less than a predetermined second reference number, the control unit may increase the second threshold value and decrease the third threshold value.

In addition, the control unit may control the communication unit to transmit a control command for the selected name to the terminal.

A vehicle according to another aspect receives a user's voice; When the user's voice is input, search for at least one name candidate corresponding to the input voice; Determining a threshold for a confidence score of the at least one name candidate based on a user pattern; And selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold.

In addition, when the phonebook information and call history information of the terminal is received, generating a user pattern based on the received phonebook information and call history information; may further include.

Further, selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold value corresponds to the input voice among the name candidates in which the reliability score exceeds a first threshold value. Selecting a name; may include.

In addition, selecting the name corresponding to the input voice among the at least one name candidate based on the determined threshold value corresponds to a case in which the number of name candidates whose reliability score exceeds the first threshold is 1 Determining whether the reliability score of the name candidate exceeds the second threshold, and when the reliability score of the name candidate exceeds the second threshold, selecting the name candidate as a name corresponding to the input voice; It may include.

In addition, if the reliability score of the name candidate whose confidence score exceeds the first threshold is less than or equal to the second threshold, requesting confirmation from the name candidate about the name candidate may be further included.

Further, selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold value, when the number of name candidates whose reliability score exceeds the first threshold value is plural. Prioritizing a plurality of name candidates based on the reliability score; Comparing the difference in reliability scores between the first and second priority name candidates and the third threshold; And selecting a name corresponding to the input voice based on the comparison result.

In addition, selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold value is a reliability score between the first-ranked name candidate and the second-ranked name candidate. And when the difference of exceeds the third threshold, selecting a name candidate whose priority is the first priority as a name corresponding to the input voice.

In addition, selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold value is a reliability score between the first-ranked name candidate and the second-ranked name candidate. If the difference of is less than or equal to the third threshold, requesting the user for confirmation of a name candidate whose priority is the first priority may include.

Also, searching for at least one name candidate may include, when phonebook information of the terminal is received, search for at least one name candidate corresponding to the input voice among the received phonebook information, and search for at least one name candidate. Determining a threshold for a confidence score based on a user pattern may include at least one of a first threshold value, a second threshold value, or a third threshold value based on the number of searched name candidates. Change; may include.

Further, changing at least one of the first threshold value, the second threshold value, or the third threshold value based on the number of searched name candidates, when the number of searched name candidates is equal to or less than a predetermined first reference number, the And increasing the second threshold value and decreasing the third threshold value.

Further, determining a threshold for a confidence score of the at least one name candidate based on a user pattern, confirms the number of categories for the searched name candidate from the phonebook information, and And changing at least one of the first threshold value, the second threshold value, or the third threshold value based on the number of categories.

Further, changing at least one of the first threshold value, the second threshold value, or the third threshold value increases the second threshold value when the number of categories is equal to or less than a predetermined second reference number, and the second threshold value is increased. 3 to decrease the threshold; may include.

In addition, it may further include; transmitting a control command for the selected name to the terminal.

According to the vehicle and its control method according to an aspect, a more accurate and fast voice recognition function may be provided, so that user convenience may be increased.

1 is a control block diagram of a vehicle according to an embodiment.
2 is a view showing the internal configuration of a vehicle according to an embodiment.
3 is a view for explaining the operation of the control unit according to an embodiment.
4 and 5 are diagrams illustrating examples of recognition results according to an embodiment.
6 is a diagram illustrating an example of a user pattern according to an embodiment.
7 and 8 are views showing an example of a screen that can be displayed on another display in one embodiment.
9 and 10 are flowcharts illustrating a method for controlling a vehicle according to an embodiment.

The same reference numerals refer to the same components throughout the specification. This specification does not describe all elements of the embodiments, and overlaps between general contents or embodiments in the technical field to which the present invention pertains are omitted. The term 'unit, module, member, block' used in the specification may be implemented by software or hardware, and according to embodiments, a plurality of 'unit, module, member, block' may be implemented as one component, It is also possible that one 'part, module, member, block' includes a plurality of components.

Throughout the specification, when a part is "connected" to another part, this includes not only a direct connection but also an indirect connection, and an indirect connection includes connecting through a wireless communication network. do.

Also, when a part “includes” a certain component, this means that other components may be further included, rather than excluding other components, unless otherwise specified.

Throughout the specification, when one member is positioned “on” another member, this includes not only the case where one member abuts another member, but also the case where another member exists between the two members.

Terms such as first and second are used to distinguish one component from other components, and the component is not limited by the above-mentioned terms.

Singular expressions include plural expressions, unless the context clearly has an exception.

In each step, the identification code is used for convenience of explanation. The identification code does not describe the order of each step, and each step can be executed differently from the specified order unless a specific order is clearly stated in the context. have.

Hereinafter, working principles and embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a control block diagram of a vehicle according to an embodiment, and FIG. 2 is a diagram showing an internal configuration of a vehicle according to an embodiment.

1 and 2, the vehicle 1 according to an embodiment includes a voice input unit 110, a communication unit 120, a speaker 151, a display 152, a control unit 130, and a storage unit 140 And it may include an input unit 160.

The vehicle 1 may receive a user's control command by voice through the voice input unit 110. The voice input unit 110 may receive sound and convert it into an electrical signal to output the sound. To this end, the voice input unit 110 may be implemented as a microphone, but is not limited thereto.

For effective voice input, the voice input unit 110 may be mounted on the headlining 64 as illustrated in FIG. 2, but the embodiment of the vehicle 1 is not limited thereto, and is mounted on the dashboard 61 It is also possible to be mounted on the steering wheel (65). In addition to this, there is no restriction on any location suitable for receiving a user's voice.

The communication unit 120 may communicate with an external server or terminal 2. The communication unit 120 may transmit and receive various information with the terminal 2. Specifically, the communication unit 120 may receive phonebook information and call history information from the terminal 2.

At this time, the phonebook information may include a name, a phone number for each name, favorite setting information for each name, and a category for each name. Here, the category by name means the type of the phone number by name, and for example, the category may be at least one of 'mobile', 'home', and 'office'.

The call history information includes name, transmission by name, reception information, transmission time by name, and reception time information by name. Here, the call history information may be arranged in chronological order from the time of the last call. In addition, the listed call history information may be listed for each transmission and transmission, respectively, and may be arranged in chronological order irrespective of whether it is originated or transmitted.

On the other hand, the terminal 2 means a device capable of performing a call function. The terminal 2 may be embodied as a computer or a portable terminal capable of accessing the vehicle 1 through a network. Here, the computer includes, for example, a laptop equipped with a web browser (WEB Browser), a desktop (desktop), a laptop (laptop), a tablet PC, a slate PC, and the like, and the portable terminal has, for example, portability and mobility. As a guaranteed wireless communication device, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication) -2000, CDMA (Code Division Multiple Access) -2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, smart phone (Smart Phone), etc. And a wearable device such as a watch, ring, bracelet, anklet, necklace, glasses, contact lens, or head-mounted-device (HMD).

The communication unit 120 may include one or more communication modules that enable communication with external devices, and may include at least one of a short-range communication module, a wired communication module, and a wireless communication module.

The short-range communication module uses a wireless communication network at a short distance, such as a Bluetooth module, an infrared communication module, a radio frequency identification (RFID) communication module, a wireless local access network (WLAN) communication module, an NFC communication module, and a Zigbee communication module. Various short-range communication modules for transmitting and receiving signals may be included.

The wired communication module includes a variety of wired communication modules such as a local area network (LAN) module, a wide area network (WAN) module, or a value added network (VAN) module, as well as a universal serial bus (USB). , HDMI (High Definition Multimedia Interface), DVI (Digital Visual Interface), RS-232 (recommended standard232), power line communication, or a plain old telephone service (POTS).

In addition to the Wi-Fi module and Wireless broadband module, the wireless communication module includes Global System for Mobile Communication (GSM), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), and universal mobile telecommunications system (UMTS). ), A wireless communication module supporting various wireless communication methods such as Time Division Multiple Access (TDMA), Long Term Evolution (LTE), and the like.

Further, the wireless communication module may include an antenna for transmitting and receiving signals, a transmitter, and a receiver.

In addition, the wireless communication module may further include a signal conversion module for modulating the digital control signal into an analog type wireless signal and demodulating the received analog type wireless signal into a digital control signal.

Also, the communication unit 120 may further include an internal communication module for communication between electronic devices inside the vehicle 1. As an internal communication protocol of the vehicle 1, a controller area network (CAN), a local interconnection network (LIN), a flexray, and an Ethernet may be used.

The communication unit 120 may exchange data with an external server using a wireless communication module, and exchange data with a mobile device using a short-range communication module or a wired communication module.

The center fascia 62, which is the central area of the dashboard 61 inside the vehicle 1, displays on the screen required to perform control of the vehicle including audio functions, video functions, navigation functions, and dialing functions ( 152) and an input unit 160 for receiving a user's control command may be provided.

The display 152 includes a cathode ray tube (CRT), a digital light processing (DLP) panel, a plasma display panel, a liquid crystal display (LCD) panel, and electroluminescence ( Electro Luminescence (EL) panel, Electrotrophic Display (EPD) panel, Electrochromic Display (ECD) panel, Light Emitting Diode (LED) panel or Organic Light Emitting Diode (OLED) ) It can be implemented as one of various display devices such as panels.

The user may input a command for controlling the vehicle 1 by operating the input unit 160. As illustrated in FIG. 2, the input unit 160 may be provided as a hard key type in an area adjacent to the display 152, and when the display 152 is implemented as a touch screen type, the display 152 is an input unit 160 ) Can also be performed.

The input unit 160 is also provided on the steering wheel, so that a driving driver can input a command related to the control of the vehicle 1 even without releasing the steering wheel.

In addition, the input unit 160 may be provided in the center input unit 43 of a jog shuttle type or a hard key type, and may include all of the components capable of inputting commands related to the control of the vehicle 1. have.

In addition, a speaker 151 for outputting sound necessary for providing a service desired by a user may be provided inside the vehicle 1. For example, the speaker 151 may be provided inside the driver's seat door 71L and the passenger seat door 71R.

The speaker 151 is a voice for providing a call function, a voice for guiding a navigation route, sound or voice included in audio / video content, a voice for providing information or a service desired by the user, and a response to a user's speech The generated voice and the like can be output.

The control unit 130 may perform overall control of the internal configuration of the vehicle 1.

The control unit 130 may control the communication unit 120 to transmit and receive various information to and from the terminal 2. Specifically, the control unit 130 may control the communication unit 120 to receive various information, including phone book information and call history information, from the terminal 2.

The controller 130 may generate a user pattern based on the phonebook information and the call history information received from the terminal 2. Specifically, the control unit 130 may generate a user pattern by extracting or collecting information necessary to understand the characteristics of the user's call history based on the phonebook information and the call history information.

At this time, the user pattern refers to information on the user's call pattern, and may specifically refer to feature information of the user's call history by time. The user pattern may include the name of the phonebook information, a category by name, a number of calls by name, a transmission time by name, a reception time by name, a transmission time for a category by name, a reception time for a category by name, and the like. In addition, the user pattern is used to determine a user's call pattern, such as whether a confidence score for each name, a confidence score for each name, at least one threshold value for each name, a recommended action by name, and a recommended action by name are performed. It may contain various information needed.

In addition, when the user's voice is input through the voice input unit 110, the control unit 130 may control the communication unit 120 to transmit a control command for the name corresponding to the input voice to the terminal 2.

At this time, the control unit 130 may control the communication unit 120 to transmit a call originating command for the name corresponding to the user's voice to the terminal 2. The control unit 130 may transmit a phone number matching the name corresponding to the user's voice, and may transmit a call origination command for the name to the terminal 2 through the communication unit 120.

To this end, the control unit 130 may search for at least one name candidate corresponding to the user's voice, and the threshold for the confidence score of the at least one name candidate based on the user pattern Can decide. The controller 130 may select a name corresponding to the input voice among the at least one name candidate based on the determined threshold. This will be described in detail later.

On the other hand, the control unit performs an operation described above using a memory (not shown) that stores data for an algorithm or a program reproducing the algorithm for controlling the operation of components in the vehicle 1 and data stored in the memory. It may be implemented as a processor (not shown). At this time, the memory and the processor may be implemented as separate chips, respectively. Alternatively, the memory and the processor may be implemented as a single chip.

The storage unit 140 may store various information such as phone book information, call history information, and user patterns of the received terminal 2.

Also, the storage unit 140 may store various information related to speech recognition, such as a sound model, a language model, and word information.

To this end, the storage unit 140 is a non-volatile memory such as cache, read only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EPMROM), and flash memory The device may be implemented as at least one of a volatile memory device such as a random access memory (RAM) or a storage medium such as a hard disk drive (HDD) or CD-ROM, but is not limited thereto. The storage unit may be a memory implemented in a separate chip from the above-described processor in relation to the controller 130, or may be implemented in a single chip from the processor.

At least one component may be added or deleted in response to the performance of the components of the vehicle 1 shown in FIG. 1. In addition, it will be readily understood by those skilled in the art that the mutual positions of the components may be changed corresponding to the performance or structure of the system.

Meanwhile, each component illustrated in FIG. 1 refers to software and / or hardware components such as a field programmable gate array (FPGA) and an application specific integrated circuit (ASIC).

Hereinafter, the operation of the specific control unit 130 will be described in detail.

3 is a view for explaining the operation of the control unit according to an embodiment.

Referring to FIG. 3, the control unit 130 according to an embodiment of the present invention is a pre-processing unit 131 performing noise removal on an input signal received from the voice input unit 110, and recognizes the input user's voice in text form. It may include a voice recognition unit 132 for outputting the speech, and a post-processing unit 133 for selecting a name corresponding to the user's voice among the recognition results.

The pre-processing unit 131 may perform noise removal on the input signal received from the voice input unit 110. Also, the pre-processing unit 131 may perform noise removal to remove noise from the input signal through various signal processing such as signal amplification.

In addition, the pre-processing unit 131 may monitor the input signal transmitted from the voice input unit 110 to determine whether a user's voice is input.

Previously, the user had to input a trigger signal, such as opening a microphone by operating a Push to Talk (PTT) button to input voice, or uttering a separate command to activate voice recognition. However, in the vehicle 1 according to an embodiment, the voice input unit 110 is always open, and the pre-processing unit 131 monitors the signal input through the voice input unit 110 in real time to determine whether a user inputs voice. I can judge.

If it is determined that the user's voice is input, the pre-processing unit 131 may perform noise removal on the input signal, and may transmit the signal-processed voice signal to the voice recognition unit 132.

The voice recognition unit 132 may recognize the voice uttered by the user from the input voice signal, and output the recognition result. The recognition result output from the speech recognition unit 132 may be a text-like speech.

To this end, the speech recognition unit 132 may include a speech recognition engine, and the speech recognition engine recognizes speech uttered by the user by applying a speech recognition algorithm to the input speech, and recognizes the recognition result. Can be created.

The speech recognition unit 132 may use an acoustic model for modeling and comparing signal characteristics of speech and a language model for modeling linguistic order relationships such as words or syllables corresponding to the recognized vocabulary. have. To this end, the acoustic model / language model DB 141 or 142 may be stored in the storage 140.

On the other hand, a language model that models a linguistic order relationship such as a word or a syllable can reduce acoustic ambiguity and reduce errors in recognition by applying the order relationship between units constituting language to units obtained in speech recognition. The language model includes a statistical language model and a model based on a Finite State Automata (FSA), and the statistical language model uses chain probabilities of words such as Unigram, Bigram, and Trigram.

The voice recognition unit 132 may use any of the above-described methods in recognizing voice. For example, an acoustic model to which the Hidden Markov model is applied may be used, or an N-best search method incorporating an acoustic model and a voice model may be used. The N-best search method can improve recognition performance by retrieving the ranks of the candidates after retrieving up to N recognition result candidates using the acoustic model and the language model.

The voice recognition unit 132 may calculate a confidence score to secure the reliability of the recognition result. The reliability score is a measure of how reliable the results of speech recognition are. For example, a phoneme or word that is a recognized result may be defined as a relative value for the probability that the word is spoken from other phonemes or words. Accordingly, the reliability score may be expressed as a value between 0 and 1, and may be expressed as a value between 0 and 10000, but is not limited thereto.

In addition, the voice recognition unit 132 may extract a name from the text-type speech output from the user's voice. In this case, the voice recognition unit 132 may search for at least one name candidate for a name corresponding to the user's voice, and the recognition result may include at least one name candidate. In addition, a reliability score for at least one name candidate may be further included.

For example, when using the N-best search method described above, the voice recognition unit 132 may search up to N name candidates. The voice recognition unit 132 may output the recognition result by evaluating the rank of the N name candidates, and calculate a reliability score for the N name candidates.

The recognition result generated by the speech recognition unit 132 may be transmitted to the post-processing unit 133. The post-processing unit 133 may select a name corresponding to the user's voice among the recognition results, and may control the communication unit 120 to transmit various control commands for the selected name to the terminal 2.

For example, the post-processing unit 133 may control the communication unit 120 to transmit a call originating command for the name corresponding to the user's voice to the terminal 2. The post-processing unit 133 may transmit a phone number matching the name corresponding to the user's voice, and may transmit a call originating command for the name to the terminal 2 through the communication unit 120.

To this end, the post-processing unit 133 may select a name corresponding to the user's voice among at least one name candidate included in the recognition result.

Specifically, the post-processing unit 153 may select a name corresponding to the user's voice among at least one name candidate included in the recognition result based on a preset threshold. At this time, the preset threshold value means a threshold value for the reliability score for the recognition result, and may be at least one.

The post-processing unit 133 may store the at least one threshold value in the reliability score DB 144, and may store the threshold value for the recognition result for each user who has entered a name or voice.

The post-processing unit 133 may select a name corresponding to the user's voice from among name candidates whose reliability score exceeds a preset first threshold, so that an operation on the selected name is performed. In this case, the first threshold value may mean a reliability score that guarantees minimum reliability for the recognition result.

The post-processing unit 133 rejects the recognition result because it cannot guarantee the reliability of the recognition result if there is no name candidate whose reliability exceeds a preset first threshold among at least one name candidate included in the recognition result ( rejection). That is, the recognition failure can be confirmed. In this case, the post-processing unit 133 may request the user to input the voice again through the display 152 or the speaker 151.

In addition, the post-processing unit 133, when there is one name candidate that exceeds the first predetermined threshold among the at least one name candidate included in the recognition result, the reliability score of the corresponding name candidate exceeds the second threshold. It is possible to determine whether or not, and a name corresponding to the user's voice can be selected based on the determination result.

At this time, the second threshold value may mean a reliability score capable of ensuring maximum reliability of the recognition result, and may be greater than the first threshold value.

Specifically, since the post-processing unit 133 guarantees high reliability when the reliability score of a name candidate whose reliability exceeds the first threshold value exceeds the second threshold value, the post-processing unit 133 applies the name candidate to the input voice. You can choose the corresponding name.

Alternatively, the post-processing unit 133 cannot confirm that the reliability of the name candidate whose reliability exceeds the first threshold is less than the second threshold, so that it does not correspond to the case where high reliability is guaranteed, so check the name candidate. The user can make a visual or audible request.

To this end, the post-processing unit 133 may control the display 152 or the speaker 151 to request the user to confirm information regarding the name candidate, and when the user checks the name candidate, The identified name candidate can be selected as a name corresponding to the user's voice.

When the number of name candidates whose reliability score exceeds the first threshold value is plural, the post-processing unit 133 may determine the priority of the plurality of name candidates based on the reliability score. The post-processing unit 133 may compare the difference in the reliability score between the name candidate having the first priority and the name candidate having the second priority and the third threshold, and corresponding to the user's voice based on the comparison result You can choose a name.

In this case, the third threshold value may mean a threshold value that guarantees reliability of the difference in the reliability score between the recognition results, and may be a threshold value having a reliability score smaller than the first threshold value and the second threshold value.

When the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate exceeds the third threshold, since the reliability of the first-ranked name candidate is guaranteed, the post-processing unit 133 May select a name candidate whose priority is the first priority as a name corresponding to the user's voice.

Alternatively, when the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is less than or equal to the third threshold, reliability of the first-ranked name candidate is not guaranteed. 133) may visually or audibly request the user to confirm the name candidate whose priority is the first priority.

In other words, the post-processing unit 133 does not immediately select the name candidate having the first priority as a name corresponding to the user's voice, but requests the user to confirm, so that the first priority is only when confirmation is made from the user. The name candidate can be selected as a name corresponding to the user's voice.

In addition, when the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is less than or equal to the third threshold, the post-processing unit 133 determines whether there are multiple categories for the first-ranked name candidate. Can be checked.

At this time, if there are not a plurality of categories for the first name candidate, that is, if the first name candidate includes only one category, the post-processing unit 133 is assigned to the first priority name candidate. You can ask the user for a confirmation, either visually or audibly.

That is, when the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is less than or equal to the third threshold, since the reliability of the first-ranked name candidate is not guaranteed, the post-processing unit In step 133, the name candidate whose priority is the first priority may not be selected as a name corresponding to the user's voice, and the user may be asked to confirm. The post-processing unit 133 may increase the accuracy of speech recognition by selecting a name candidate that is the first priority as a name corresponding to the user's voice only when confirmation is made from the user.

If there are multiple categories for the first name candidate, that is, the first name candidate includes only a plurality of categories, the post-processing unit 133 determines the probability value of each of the plurality of categories based on the stored user pattern. Can be.

In this case, the post-processing unit 133 may determine a probability value of each of the plurality of categories based on at least one of a transmission time for a category by name of a user pattern or a reception time for a category by name.

The post-processing unit 133 may select a category having a maximum probability value based on a plurality of category-specific probability values, and check whether the probability value of the selected category exceeds a predetermined reliability score.

When the probability value of the selected category exceeds a predetermined reliability score, the post-processing unit 133 may perform control for the selected category. That is, the post-processing unit 133 may transmit a call originating command for a phone number matching the selected category for the first rank name candidate.

In this case, since the post-processing unit 133 uses a user pattern reflecting the call characteristics of the user, a more accurate call originating command can be transmitted. Therefore, in providing a call function using voice recognition, since a more accurate call function may be provided, usability and convenience may be increased.

Alternatively, if the probability value of the selected category is equal to or less than a predetermined reliability score, the post-processing unit 133 may request the user to select a category. In this case, the post-processing unit 133 may request the user to select a category by displaying a predetermined number of category candidates to the user without directly controlling the selected category.

Thereafter, when a selection of a category is received from the user, the post-processing unit 133 may perform control for the selected category. That is, the post-processing unit 133 may transmit a call originating command for a phone number matching the selected category for the first rank name candidate.

Through this, a more accurate user voice recognition function may be provided, and a user's convenience may be increased because a user's voice recognition success rate is increased.

In addition, the post-processing unit 133 may change a predetermined threshold value. Specifically, the post-processing unit 133 may determine at least one threshold value based on the user pattern related to the user's call pattern.

To this end, the post-processing unit 133 may generate a user pattern based on the phonebook information and call history information of the terminal, and may store the user pattern in the user pattern DB 143.

Specifically, the post-processing unit 133 may generate a user pattern by extracting or collecting information necessary to understand the characteristics of the user's call history based on the phonebook information and the call history information.

At this time, the user pattern refers to information on the user's call pattern, and may specifically refer to feature information of the user's call history by time. The user pattern may include the name of the phonebook information, a category by name, a number of calls by name, a transmission time by name, a reception time by name, a transmission time for a category by name, and a reception time for a category by name. In addition, the user pattern is used to determine a user's call pattern, such as whether a confidence score for each name, a confidence score for each name, at least one threshold value for each name, a recommended action by name, and a recommended action by name are performed. It may contain various information needed.

Further, the post-processing unit 133 may search for at least one name candidate corresponding to the user's voice among the phonebook information of the terminal 2, and may change at least one threshold value based on the number of searched name candidates. . That is, the post-processing unit 133 may change at least one of the first threshold value, the second threshold value, or the third threshold value.

Specifically, when the number of searched name candidates is equal to or less than the predetermined first reference number, the post-processing unit 133 may increase the second threshold value. If the number of searched name candidates is less than or equal to a predetermined first reference number, it may mean that there is no similar name in the phone book. In this case, the probability that an exact name corresponding to the user's voice is searched may be higher than if there are many similar names in the phone book. That is, it may be an environment with a high recognition rate. Accordingly, the post-processing unit 133 may reduce the number of name candidates included in the recognition result by increasing the second threshold value, and control the selected name to be performed immediately. Therefore, user usability can be increased.

In addition, when the number of searched name candidates is equal to or less than the predetermined first reference number, the post-processing unit 133 may decrease the third threshold value. In this case, even if the difference in the reliability score between the recognition results is small, since the first candidate name can be selected without requesting confirmation to the user, the user's usability can be increased.

In addition, when the number of searched name candidates exceeds a predetermined first reference number, the post-processing unit 133 may decrease at least one of the first threshold value or the second threshold value. When the number of searched name candidates exceeds a predetermined first reference number, it may mean that there are many similar names in the phone book. In this case, the probability that an accurate name corresponding to a user's voice is searched may be lower than when there are few similar names in the phone book. That is, it may be an environment with low recognition rate.

Accordingly, the post-processing unit 133 may increase the number of name candidates included in the recognition result by reducing at least one of the first threshold value or the second threshold value, and prevent control of the selected name from being performed immediately. . Therefore, the accuracy and recognition rate of speech recognition can be increased.

In addition, when the number of searched name candidates exceeds a predetermined first reference number, the post-processing unit 133 may increase the third threshold value. In this case, even if the difference in the reliability score between the recognition results is large, a request for confirmation is made to the user, so the accuracy and recognition rate of speech recognition may be increased.

In addition, the post-processing unit 133 may check the number of categories for the name candidates retrieved from the phonebook information of the terminal 2, and may change at least one threshold value based on the number of confirmed categories.

Specifically, when the number of categories for the searched name candidate is less than or equal to a predetermined second reference number, the post-processing unit 133 may increase at least one of the first threshold value or the second threshold value.

When the number of categories for the searched name candidate is less than or equal to the predetermined second reference number, it may mean that the number of phone numbers matching the name is small. In this case, the probability of providing a function that meets the needs of the user may be higher than when there are many types of phone numbers matching the name. That is, it may be an environment with a high recognition rate.

Therefore, the post-processing unit 133 may reduce the number of name candidates included in the recognition result by increasing at least one of the first threshold value or the second threshold value, and allow control of the selected name to be performed immediately. have. Therefore, user convenience may be increased.

In addition, when the number of categories for the searched name candidate is less than or equal to a predetermined second reference number, the post-processing unit 133 may decrease the third threshold value. In this case, even if the difference in the reliability score between the recognition results is small, the user can select the first name candidate without the user's request for confirmation, thereby increasing user convenience.

Alternatively, when the number of categories for the searched name candidate exceeds a predetermined second reference number, the post-processing unit 133 may decrease at least one of the first threshold value or the second threshold value.

When the number of categories for the searched name candidate exceeds a predetermined second reference number, it may mean that there are many types of phone numbers matching the name. In this case, the probability of providing a function that meets the user's needs may be lower than when the number of phone numbers matching the name is small. That is, it may be an environment with low recognition rate.

Accordingly, the post-processing unit 133 may increase the number of name candidates included in the recognition result by reducing at least one of the first threshold value or the second threshold value. The post-processing unit 133 may increase the accuracy and recognition rate of speech recognition by allowing the user's confirmation request not to be performed immediately for the selected name.

In addition, when the number of categories for the searched name candidate exceeds a predetermined second reference number, the post-processing unit 133 may increase the third threshold value. In this case, even if the difference in the reliability score between the recognition results is large, a request for confirmation of the user is performed, and a name candidate confirmed from the user can be selected. Therefore, the accuracy and recognition rate of speech recognition can be increased.

The post-processing unit 133 may update the threshold value by storing the changed at least one threshold value in the reliability score DB 144. In addition, the post-processing unit 133 may update the user pattern by generating a user pattern for the changed threshold value or by changing the threshold value included in the stored user pattern to the changed threshold value.

Meanwhile, the pre-processing unit 131, the voice recognition unit 132, and the post-processing unit 133 may include separate memories and processors, and may share the memory and the processor with each other.

In addition, a plurality of memory and processors may be used. In this case, a plurality of memories and processors may be integrated on one chip, or may be provided in physically separate locations.

At least one component may be added or deleted with respect to the performance of the components of the control unit 130 illustrated in FIG. 3. In addition, it will be readily understood by those skilled in the art that the mutual positions of the components may be changed corresponding to the performance or structure of the system.

4 is a diagram illustrating an example of a recognition result according to an embodiment.

Referring to FIG. 4, when the user utters "Call James," the controller 130 according to an embodiment may extract the name James from the voice of the user.

The controller 130 may search for at least one name candidate for the user's voice 'James', and calculate reliability scores of each name candidate.

The control unit 130 may select a name corresponding to the user's voice among name candidates (James, Jason, Jameson, Jade, Jane, Mason, Jeremy) whose reliability score exceeds a preset first threshold value T1, Name candidates (Jade, Johnson, John) having a reliability score below the first threshold T1 may be excluded from the recognition result. At this time, the first threshold value T1 may be a value between the reliability scores X8 and X7.

The controller 130 has a plurality of name candidates having a reliability score exceeding a preset first threshold value T1, so that the priority according to the reliability score is the first name candidate and the second priority name It can be confirmed whether the difference in the reliability score of the candidate (Jason) exceeds the third threshold (T3).

When the difference in the reliability score between the first-ranked name candidate James and the second-ranked name candidate Jason exceeds the third threshold T3, the control unit 130 first-ranked name candidate (James) ) May determine whether to request confirmation from the user according to whether the reliability exceeds the second threshold T2.

When the reliability of the first name candidate (James) exceeds the second threshold value (T2), since reliability is guaranteed, the controller 130 corresponds to the first name candidate (James) corresponding to the user's voice. You can choose by name.

The control unit 130 may transmit a call to the communication unit 120 by sending a phone call command for the selected name (James) so that the call is sent to the number matching 'James'.

Alternatively, if the reliability of the first name candidate (James) is less than or equal to the second threshold value (T2), since the reliability is not guaranteed, the controller 130 checks for the first-ranked name candidate (James). You can ask.

If the difference in the reliability score between the first name candidate (James) and the second ranking name candidate (Jason) is equal to or less than the third threshold value (T3), the control unit 130 is assigned to the first name candidate (James). You can ask the user for a confirmation visually or audibly. Through this, if the reliability is not guaranteed, the accuracy of speech recognition may be increased by appropriately requesting confirmation from the user.

5 is a diagram illustrating an example of a recognition result according to an embodiment.

Referring to FIG. 5, when the user utters "Call James," the controller 130 according to an embodiment may extract the name James from the voice of the user. The controller 130 may search for at least one name candidate for the user's voice “James” and may calculate reliability scores of each name candidate.

The control unit 130 has the number of at least one name candidate corresponding to 'James' among the phonebook information of the terminal 2 is equal to or less than a predetermined first reference number, or a second reference number having a predetermined number of categories for the searched name candidates In the following cases, the first threshold value T1 and the second threshold value T2 may be increased and the third threshold value T3 may be decreased.

The controller 130 may increase the number of excluded name candidates from 2 (Johnson, John) to 5 (Mason, Jeremy, Jade, Johnson, John) by increasing the first threshold T1. That is, the control unit 130 may reduce the number of name candidates included in the consideration range.

In addition, the control unit 130 may increase the second threshold value T2 so that control of the selected name can be performed immediately. That is, when the user is in an environment where the recognition rate is high, the controller 130 may control the selected name without user confirmation.

In addition, the control unit 130 reduces the third threshold value T3 so that even if the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is small, control of the selected name can be performed immediately. can do. In this case, even if the difference in the reliability score between the recognition results is small, since the first candidate name can be selected without requesting confirmation to the user, the user's usability can be increased.

Through this, the controller 130 can detect a case in which the user is in an environment with a relatively high recognition rate, and may increase the processing speed of speech recognition. Therefore, more accurate and faster speech recognition can be implemented.

6 is a diagram illustrating an example of a user pattern according to an embodiment.

Referring to FIG. 6, when a user inputs a voice, the controller 130 according to an embodiment may generate a user pattern including a selected name corresponding to the input voice.

The controller 130 may include and store information utilized as a control basis in the user pattern while the user uses the voice recognition function. That is, the controller 130 may include information used to select a name corresponding to the input user voice in the user pattern.

For example, the controller 130 may include category information (mobile), a reliability score (5100), a first threshold value (3000), a second threshold value (5000), and a third threshold value (500) for the selected name (Hong Gil-dong). ) May be stored as a user pattern, and a user pattern including a recommended action (direct call) and whether or not to be performed (eg) may be generated and stored.

In this case, the recommended action may indicate the type of the control command generated by the control unit 130, and when a call originating command is generated for the selected name without a user's request for confirmation (Direct Call), confirms after requesting confirmation from the user It may include a case in which a phone call command is generated for a given name (Confirmation Call) and a case in which a phone call command is generated for a selected name after requesting a user's selection among a plurality of candidates (N-best Call).

In addition, the control unit 130 may generate a user pattern for the selected name (Hong Gil-dong) every time the user uses the voice recognition function and make it a database, which can be used for later threshold determination. Therefore, it is possible to provide a speech recognition function suitable for an individual user's environment.

7 is a view showing an example of a screen that can be displayed on another display in one embodiment.

Referring to FIG. 7, when a voice (“Call James”) is input from a user, the control unit 130 according to an embodiment may use at least one name candidate 152a corresponding to the name of the input voice (James). You can ask the user to choose a name by displaying.

The user may select one of the displayed name candidates 152a, and upon receiving a name corresponding to the user's voice from the user, the controller 130 may transmit a control command for the selected name to the communication unit 120. Specifically, the control unit 130 may transmit a call origination command for the selected name to the communication unit 120.

In addition, the controller 130 may request selection of a category for a name corresponding to a user's voice as well as selection for a name.

Specifically, if the probability value of a category for a name corresponding to a user's voice is equal to or less than a predetermined reliability score, the controller 130 may request the user to select a category. In this case, the controller 130 may request the user to select a category by displaying a predetermined number of category candidates to the user.

Thereafter, when a selection of a category is received from the user, the control unit 130 may perform control for the selected category. That is, the controller 130 may transmit a call originating command for a phone number matching the selected category.

8 is a view showing an example of a screen that can be displayed on another display in one embodiment.

Referring to FIG. 8, the controller 130 according to an embodiment may visually request confirmation of a name candidate corresponding to a voice (“Call James”) from a user. The controller 130 may request the user to confirm the selected name candidate by displaying information on the selected name candidate (James). At this time, information displayed as information on the selected name candidate (James) may include a name (James), a category (Mobile), and a phone number (01234567890).

When the displayed name corresponds to the voice input name, the user may input a confirmation of the name candidate through the confirmation whether interfaces 152b and 152c displayed on the display 152.

When the name candidate is confirmed by the user, the identified name candidate may be selected as a name corresponding to the user's voice, and control of the selected name may be performed. That is, the controller 130 may transmit a call origination command for a phone number matching the selected name.

9 is a flowchart illustrating a method for controlling a vehicle according to an embodiment.

Referring to FIG. 9, the vehicle 1 according to an embodiment may check whether a voice is input from a user (701).

When a voice is input from the user (YES in 701), the vehicle 1 may recognize the voice and search for a name corresponding to the input voice (702). Specifically, the vehicle 1 may search for at least one name candidate corresponding to the input voice.

The vehicle 1 may calculate a reliability score for at least one name candidate (703). For example, when using the N-best search method, the vehicle 1 can search up to N name candidates. The vehicle 1 can output the recognition result by evaluating the rank of N name candidates, and can calculate a reliability score for the N name candidates.

Thereafter, the vehicle 1 may check whether the reliability score for each name candidate exceeds a preset first threshold, and among the searched name candidates, a name candidate whose reliability score exceeds the first threshold It can be checked whether it exists (704). In this case, the first threshold value may mean a reliability score that guarantees minimum reliability for the recognition result.

If there is a name candidate whose reliability score exceeds the first threshold among the searched name candidates (YES in 704), the vehicle 1 has a reliability score among the searched name candidates that exceeds a preset first threshold. A name candidate can be extracted, and it can be checked whether the number of extracted name candidates is one (705).

In addition, when there is one name candidate whose reliability exceeds a preset first threshold among at least one name candidate included in the recognition result (example of 705), the vehicle 1 corresponds to the searched name candidate corresponding to the user's voice. It can be selected by name (706), and it can be determined whether the reliability score of the selected name candidate exceeds the second threshold (707).

At this time, the second threshold value may mean a reliability score capable of ensuring maximum reliability of the recognition result, and may be greater than the first threshold value.

When the reliability score of the selected name candidate exceeds the second threshold (YES in 707), the vehicle 1 may perform control of the selected name (708). Specifically, the vehicle 1 may perform a call operation for the selected name by transmitting a call control command for the selected name to the terminal 2.

Thereafter, the vehicle 1 may check whether the operation according to the control command for the selected name is canceled (709), and when the operation is canceled by the user (YES in 709), the reliability score DB may be updated ( 710).

Specifically, the vehicle 1 may change the reliability score for the recognition result stored in the reliability score DB. For example, when the operation is canceled by the user, the vehicle 1 may determine that there is an error in the recognition result, and reduce the reliability score for the recognition result. Through this, the vehicle 1 can improve accuracy for speech recognition.

As another example, when the reliability score of the selected name candidate is less than or equal to the second threshold (No in 707), since reliability is not guaranteed, the vehicle 1 may request confirmation of the user for the selected name (714).

When confirmation of the corresponding name candidate is made by the user, the vehicle 1 may select the identified name candidate as a name corresponding to the user's voice. Thereafter, the vehicle 1 performs operations 708 to 710, and is the same as described above.

As another example, if there is no name candidate whose reliability score exceeds the first threshold among the searched name candidates (No in 704), the reliability of the recognition result cannot be guaranteed, so the vehicle 1 rejects the recognition result ( rejection). That is, the recognition failure can be confirmed. In this case, the vehicle 1 may request the user to input the voice again through the display 152 or the speaker 151.

As another example, if there is a plurality of name candidates whose reliability exceeds a preset first threshold among at least one name candidate included in the recognition result (No in 705),

The vehicle 1 may determine the priority of the plurality of name candidates based on the reliability score, and whether or not the difference in the reliability scores between the first and second ranks among the plurality of name candidates exceeds the third threshold value. It can be confirmed (712).

In this case, the third threshold value may mean a threshold value that guarantees reliability of the difference in the reliability score between the recognition results, and may be a threshold value having a reliability score smaller than the first threshold value and the second threshold value.

If the difference in the reliability score between the first and second ranks among the plurality of name candidates exceeds the third threshold value (YES in 712), the reliability of the first name candidate, which is the first rank, is guaranteed, so the vehicle 1 May select a name candidate having a first priority as a name corresponding to the user's voice (713). Thereafter, the vehicle 1 performs operations 708 to 710, and is the same as described above.

As another example, the operation of the vehicle 1 when the difference in the reliability score between the first rank and the second rank among the plurality of name candidates is less than or equal to the third threshold (No in 712) will be described with reference to FIG. 10.

10 is a flowchart illustrating a method for controlling a vehicle according to an embodiment.

Referring to FIG. 10, when a difference in a reliability score between a first rank and a second rank among a plurality of name candidates is equal to or less than a third threshold (No in FIGS. 9 and 712), the vehicle 1 according to an embodiment It may be checked whether there are a plurality of categories included in the first rank among the name candidates (801).

At this time, the category refers to the type of the phone number, and may be at least one of 'mobile', 'home', and 'office'.

If there are multiple categories included in the first rank among the name candidates (YES in 801), the vehicle 1 may select a category based on the user pattern (802).

Specifically, the vehicle 1 may determine a probability value of each of a plurality of categories based on at least one of a transmission time for a category by name of a user pattern or a reception time for a category by name. The vehicle 1 may select a category having a maximum probability value based on a plurality of category-specific probability values.

Then, the vehicle 1 may check whether the probability value of the selected category exceeds the fourth threshold value (803). At this time, the fourth threshold value may mean a predetermined reliability score, and may be stored in the reliability score DB 144.

When the probability value of the selected category exceeds the fourth threshold value (YES in 803), the vehicle 1 may perform control for the selected category (804). That is, the vehicle 1 may transmit a call originating command for a phone number matching the selected category for the first rank name candidate.

In this case, since the vehicle 1 uses a user pattern reflecting the call characteristics of the user, a more accurate call originating command can be transmitted. Therefore, in providing a call function using voice recognition, since a more accurate call function may be provided, usability and convenience may be increased.

As another example, when the probability value of the selected category is less than or equal to the fourth threshold (No in 803), the vehicle 1 may request the user to select a category (806). In this case, the vehicle 1 may request the user to select a category by displaying a predetermined number of category candidates to the user without directly controlling the selected category.

The vehicle 1 may check whether a selection of a category is received from the user (807), and when a selection of a category is received from the user (YES in 807), the vehicle 1 may perform control for the selected category. have. That is, the vehicle 1 may transmit a call originating command for a phone number matching the selected category for the first rank name candidate.

Through this, a more accurate user voice recognition function may be provided, and a user's convenience may be increased because a user's voice recognition success rate is increased.

As another example, if there are not a plurality of categories included in the first rank among the name candidates (No in 801), the vehicle 1 may request the user to select a category (805). In this case, the vehicle 1 may request the user to select a category by displaying a predetermined number of category candidates to the user without directly controlling the selected category.

Then, when a selection of a category is received from the user, the vehicle 1 may perform control for the selected category. That is, the vehicle 1 may transmit a call originating command for a phone number matching the selected category for the first rank name candidate.

Through this, a more accurate user voice recognition function may be provided, and a user's convenience may be increased because a user's voice recognition success rate is increased.

Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium that stores instructions executable by a computer. Instructions may be stored in the form of program code, and when executed by a processor, may generate program modules to perform operations of the disclosed embodiments. The recording medium may be embodied as a computer-readable recording medium.

The computer-readable recording medium includes all kinds of recording media storing instructions that can be read by a computer. For example, there may be a read only memory (ROM), a random access memory (RAM), a magnetic tape, a magnetic disk, a flash memory, and an optical data storage device.

As described above, the disclosed embodiments have been described with reference to the accompanying drawings. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention may be practiced in different forms from the disclosed embodiments without changing the technical spirit or essential features of the present invention. The disclosed embodiments are illustrative and should not be construed as limiting.

110: voice input unit
120: communication unit
130: control unit
140: storage
151: speaker
152: display
160: input unit

Claims (26)

A communication unit communicating with a terminal providing a call function;
A voice input unit that receives a user's voice;
A storage unit that stores a user pattern related to the user's call pattern; And
When the voice of the user is input, at least one name candidate corresponding to the input voice is searched, and a threshold for a confidence score of the at least one name candidate is based on the user pattern. And a controller which determines and selects a name corresponding to the input voice among the at least one name candidate based on the determined threshold. According to claim 1,
The control unit,
A vehicle that generates a user pattern based on the received phonebook information and call history information when the phonebook information and call history information of the terminal are received. According to claim 1,
The control unit,
A vehicle that selects a name corresponding to the input voice among name candidates whose reliability score exceeds a first threshold. According to claim 3,
The control unit,
When the number of name candidates whose reliability score exceeds the first threshold is 1, it is determined whether the reliability score of the name candidate exceeds the second threshold, and the reliability score of the name candidate is the second threshold If it exceeds, the vehicle selecting the name candidate as a name corresponding to the input voice. The method of claim 4,
The control unit,
If the reliability score of the name candidate whose reliability score exceeds the first threshold is less than or equal to the second threshold, the vehicle requesting confirmation from the name candidate to the user. According to claim 3,
If the number of name candidates whose reliability score exceeds the first threshold is plural, the priority of the plurality of name candidates is determined based on the reliability score, and the name candidate and the priority whose priority is the first priority A vehicle that compares a difference in reliability score between a second-ranked name candidate and a third threshold value and selects a name corresponding to the input voice based on the comparison result. The method of claim 6,
The control unit,
When the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate exceeds the third threshold value, the first-ranked name candidate corresponds to the input voice. Vehicle to choose by name to say. The method of claim 6,
If the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is less than or equal to the third threshold, the user is requested to confirm the first-ranked name candidate. Vehicle. According to claim 1,
The control unit,
When the phonebook information of the terminal is received, at least one name candidate corresponding to the input voice is searched among the received phonebook information, and a first threshold value, a second threshold value, or Vehicles that change at least one of the third thresholds. The method of claim 9,
The control unit,
A vehicle that increases the second threshold value and decreases the third threshold value when the number of searched name candidates is equal to or less than a predetermined first reference number. The method of claim 9,
The control unit,
A vehicle that checks the number of categories for the searched name candidate from the phonebook information and changes at least one of a first threshold value, a second threshold value, or a third threshold value based on the number of categories. The method of claim 11,
The control unit,
A vehicle that increases the second threshold value and decreases the third threshold value when the number of categories is equal to or less than a predetermined second reference number. According to claim 1,
The control unit,
A vehicle that controls the communication unit to transmit a control command for the selected name to the terminal. Receiving the user's voice;
When the user's voice is input, search for at least one name candidate corresponding to the input voice;
Determining a threshold for a confidence score of the at least one name candidate based on a user pattern;
And selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold. The method of claim 14,
And generating a user pattern based on the received phonebook information and call history information when the phonebook information and call history information of the terminal are received. The method of claim 14,
Selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold,
And selecting a name corresponding to the input voice among name candidates in which the reliability score exceeds a first threshold value. The method of claim 16,
Selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold,
When the number of name candidates whose reliability score exceeds the first threshold is 1, it is determined whether the reliability score of the name candidate exceeds the second threshold, and the reliability score of the name candidate is the second threshold If it exceeds, selecting the name candidate as a name corresponding to the input voice; Vehicle control method comprising a. The method of claim 17,
And if the reliability score of the name candidate whose reliability score exceeds the first threshold is less than or equal to the second threshold, requesting the user to confirm the name candidate. The method of claim 16,
Selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold,
Determining the priority of the plurality of name candidates based on the reliability score when the number of name candidates having the reliability score exceeds a first threshold is plural;
Comparing the difference in reliability scores between the first and second priority name candidates and the third threshold; And
And selecting a name corresponding to the input voice based on the comparison result. The method of claim 19,
Selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold,
When the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate exceeds the third threshold value, the first-ranked name candidate corresponds to the input voice. How to control the vehicle, including; to choose by name. The method of claim 19,
Selecting a name corresponding to the input voice among the at least one name candidate based on the determined threshold,
If the difference in the reliability score between the first-ranked name candidate and the second-ranked name candidate is less than or equal to the third threshold, the user is requested to confirm the first-ranked name candidate. The control method of the vehicle comprising a. The method of claim 14,
Searching for at least one name candidate,
When the phonebook information of the terminal is received, at least one name candidate corresponding to the input voice is searched among the received phonebook information,
Determining a threshold for a confidence score of the at least one name candidate based on a user pattern,
And changing at least one of a first threshold value, a second threshold value, or a third threshold value based on the number of searched name candidates. The method of claim 22,
Changing at least one of the first threshold value, the second threshold value, or the third threshold value based on the number of searched name candidates,
Increasing the second threshold value and decreasing the third threshold value when the number of searched candidates is less than or equal to a predetermined first reference number; Control method of a vehicle comprising a. The method of claim 22,
Determining a threshold for a confidence score of the at least one name candidate based on a user pattern,
Vehicle checking the number of categories for the searched name candidate from the phonebook information and changing at least one of a first threshold value, a second threshold value, or a third threshold value based on the number of categories. Control method. The method of claim 24,
Changing at least one of the first threshold value, the second threshold value, or the third threshold value,
If the number of categories is less than or equal to a predetermined second reference number, increasing the second threshold value and decreasing the third threshold value; Control method of a vehicle comprising a. The method of claim 14,
And transmitting a control command for the selected name to the terminal.

Images