JP4705398B2 - Voice guidance device, control method and program for voice guidance device - Google Patents

Description

  The present invention relates to a voice guidance device used for route guidance or the like of a navigation device, and relates to a voice guidance device capable of registering a voice desired by a user, not limited to an existing voice.

  In recent years, navigation devices that have rapidly become popular have a database obtained by digitizing roads, place names, buildings, etc. on a map stored in a storage medium such as a CD-ROM, and a destination is set based on the digital map database. The taxiway is searched and calculated according to the road database stored in the device in advance, and the obtained vehicle and the vehicle position calculated sequentially from navigation data such as GPS, gyroscope, vehicle speed pulse, etc. are displayed on the screen. In addition, the optimum route to the designated destination is also calculated and guided (navigated) by screen display or synthesized voice.

Conventionally, by inputting a user's desired voice (hereinafter referred to as “input voice”) from the input device as a synthesized voice in such guidance guidance, an existing voice registered in advance (hereinafter referred to as “existing voice”) can be freely set. There is a voice guidance device that can be changed. This input voice is, for example, the existing voice “soon to the left” and “turn right after paying attention to the oncoming car at the next intersection” (see Patent Document 2). Must have the same content as the audio. That is, the existing voice and the input voice need to correspond.
JP 2000-258180 A JP-A-9-325043

  However, in the prior art, even when the existing voice does not correspond to the input voice, there is usually no recording limitation for the input voice, and any voice input and registered from the input device is free. Was registered. For example, even if a right / left turn is entered in the opposite direction so that “Please turn right at the next intersection” should be registered as “Please turn left at the next intersection”.

  In a device without such checking means, if the user registers the desired voice and is set to execute the guidance guidance by this, if the wrong content is recorded at the time of recording, the wrong content will remain as it is Guidance guidance is made at. To put it extremely, it may be possible to guide a right or left turn in the opposite direction as described above, and accurate guidance guidance may not be possible. Further, such an erroneous voice guidance is naturally inconsistent with the screen display, which may confuse the user.

  The present invention solves the above-described problems, and an object of the present invention is to accurately determine whether or not a user input voice input from a predetermined input device corresponds to an existing voice. An object of the present invention is to provide a voice guidance device that can be registered, a control method and a control program for the voice guidance device.

In order to achieve the above object, the invention of claim 1 is a navigation system comprising a voice recognition unit for recognizing an input voice inputted from a predetermined input means, and a voice registration unit for registering the voice as a guidance voice. A voice guidance device used for route guidance or the like in the device includes a voice determination unit that determines whether the input voice corresponds to an already registered existing voice, and a keyword is set for each existing voice in the existing voice And the voice determination unit includes means for determining that the keyword is included in the input voice, and the voice registration unit receives the input voice when the keyword is included in the input voice. It is characterized by registering.

The invention according to claim 4 captures the invention according to claim 1 from the viewpoint of a method, wherein the control unit recognizes an input voice inputted from a predetermined input means, and guides the voice. In the control method of the voice guidance device used for route guidance guidance in the navigation device, which executes voice registration processing registered as voice, the control unit determines whether the input voice corresponds to the existing voice that has already been registered. A keyword is set for each existing voice in the existing voice, and the voice judgment process determines that the keyword is included in the input voice. In the registration process, when the keyword is included in the input voice, the input voice is registered by the control unit.

The invention of claim 6 is the which is assumed to the invention of claim 1, 4 in terms of a computer program, a control unit, to recognize the voice input from the predetermined input unit, the guidance voice the voice In a control program for a voice guidance device used for route guidance in a navigation device, which functions as a voice registration unit for registration , the control unit determines whether the input voice corresponds to an existing voice that has already been registered. A keyword is set for each existing voice in the existing voice, and the voice determination means determines that the keyword is included in the input voice. wherein the voice registration means, when the keyword is included in the input speech is for registering the input speech To.

  In the aspect as described above, the input voice input from the input means such as the microphone is analyzed, it is determined whether or not the analyzed voice corresponds to the existing voice, and the voice corresponding to the existing voice is registered. Thus, it is possible to register the input voice accurately that matches the content of the existing voice.

In addition , since the comparison between the existing voice and the input voice is performed by pattern matching whether or not a predetermined keyword included in the existing voice is included in the input voice, it is allowed to match the contents of the existing voice and the input voice. By setting the greatest common divisor condition as a keyword, the degree of freedom in selecting input speech is increased, and it becomes possible to register input speech that is compatible with various user preferences and ensures accuracy. .

According to a second aspect of the present invention, in the first aspect of the invention, the voice guidance by the existing voice and the voice guidance by the input voice are provided with a voice guidance switching unit that can be switched at any time by a user operation. To do.

The invention of claim 5 captures the invention of claim 2 from the viewpoint of a method. In the invention of claim 4 , the voice guidance by the existing voice and the voice guidance by the input voice are user operations. Thus, a voice guidance switching process that enables switching at all times is executed.

The invention of claim 7 is the one in which the inventions of claims 2 and 5 are captured from the viewpoint of a computer program. In the invention of claim 6 , the voice guidance by the existing voice and the voice guidance by the input voice are The control unit is made to function as voice guidance switching means that can always be switched by a user operation.

  In the above-described aspect, the user can select and register the voice guidance according to the driving situation at any time by enabling the change of the existing voice in response to the change request of the user at any time such as during the navigation process. Will be able to.

The invention of claim 3 is the invention according to claim 1 or 2 , characterized in that the input means includes not only a sound collector but also an external storage means for storing an audio signal.

  In the above aspect, input voice input is not limited to a microphone, and for example, input by an external storage medium such as a flash memory card or a USB memory, a portable terminal, or the like can be performed. User convenience can be improved.

  As described above, according to the present invention, a voice guidance device capable of accurate voice registration by determining whether or not a user's input voice input from a predetermined input device corresponds to an existing voice, A control method and control program for a voice guidance device can be provided.

  Next, a voice guidance apparatus (hereinafter referred to as “the present apparatus”) of the present invention, a corresponding control method, and a best embodiment for implementing a control program will be described with reference to the drawings. In the following embodiment, a case where the voice guidance device of the present invention is installed in a navigation system will be described as an example. In addition, the assumptions common to the prior art already described are not reiterated.

[1. Constitution〕
The present embodiment is a navigation system represented by a vehicle-mounted type, but can be grasped as a voice guidance device of the present invention mounted on a navigation device, and a hardware configuration example of this device is shown in FIG. 1 is an absolute position / orientation detection unit represented by GPS, and includes an GPS antenna, a receiver, etc., and an absolute position / orientation detection unit 1 that detects the position of the vehicle. Used as Reference numeral 2 denotes a relative orientation detection unit that uses a gyro or the like, and is used as the relative orientation detection unit 2 that detects the traveling direction of the host vehicle.

  Reference numeral 3 denotes a vehicle speed detection unit for processing vehicle speed pulses obtained from the vehicle. 5 to 8 are memories, and 5 of these is a ROM for storing a program to be executed by the control unit 4. A dynamic RAM (DRAM) 6 stores data processed by the control unit 4. Reference numeral 7 denotes a static RAM (SRAM) that retains memory contents such as setting from power OFF to ON again, and is backed up by a lithium battery or the like while the main power is OFF. Reference numeral 8 denotes a VideoRAM (VRAM) for the display unit 10.

  A user interface unit 9 connects the display unit 10 or the input unit 11 and the control unit 4. Reference numeral 10 denotes a display unit such as a TFT liquid crystal display panel that displays a map, a menu, and the like. Reference numeral 11 denotes an input unit for inputting various information by the user, such as a push button switch, a remote control unit, and a touch panel.

  A disk control unit 12 reads a database for guidance guidance, particularly voice guidance for voice guidance, facility search, route search, map display, and the like from a storage medium (eg, CD-ROM, DVD-ROM, HDD, etc.). is there.

  Here, the database for guidance guidance, especially voice guidance, will be explained in detail. This database is a standard voice guidance such as when turning right or left, or when the vehicle reaches the destination, and each guidance on the map. It stores existing voice data composed of point-specific voice guidance. In the present embodiment, of the existing voice data, changeable voice data is collected as a changeable existing voice list. In addition, a keyword for each existing voice data that can be changed is stored in the database as a keyword template. For example, as shown in FIG. 4, the standard existing voice data “1” at the next intersection is “1”, and this keyword is composed of “tsugi” and “migi” or “usset”. .

  A communication device and processing unit 13 receives FM broadcast waves, extracts desired data such as VICS, and is equipped with a mobile phone, PDA, or Bluetooth (trademark) via a network device such as a wired or wireless modem. It is a part which communicates with the terminal etc. (henceforth the portable terminal 21) which performed, and transmits / receives data with the portable terminal 21.

  Reference numeral 14 denotes an external storage device control unit which receives data related to navigation in the control unit 4 or input voice data in voice guidance from an external storage device 20 such as a removable flash memory card or USB memory known in various standards. This is the part to read. Reference numeral 15 denotes a microphone that collects input sound.

  Reference numeral 4 denotes a control unit (main CPU and its peripheral circuits) that controls the entire system, and performs information processing including control of other units. The control unit 4 realizes and executes a function and action as described later by a predetermined control program, the change receiving unit 41, the signal processing unit 42, the voice recognition unit 43, the keyword determination unit 44, and the voice registration. It plays a role as part 45.

  Specifically, the change receiving unit 41 is a unit that receives a request for changing an existing voice through the input unit 11 from the user. By accepting the change request here, the present apparatus enters the existing voice change mode, and the microphone 15 or the external storage device control unit 14 as well as the communication device and the processing unit 13 enter an input standby state.

  The change receiving unit 41 is a means for presenting a changeable existing voice list to the user and requesting selection of which of the existing voice data to change. Thereby, the existing voice desired to be changed is specified, and the keyword determination unit 44 described later is notified of which data of the keyword template the existing voice is.

  The signal processing unit 42 is a part that converts the analog audio signal input from the microphone 15 into a digital signal and records the processed audio data in the memory M. When the audio data is provided from the external storage device 20 or the portable terminal 21, the signal processing unit 42 does not perform processing, and the external storage device control unit 14 converts the audio data stored in the external storage device 20. The voice data transmitted from the portable terminal 21 is received by the communication device and the processing unit 13, and stored in the memory M.

  The voice recognition unit 43 is a part that reads the recorded voice data from the memory M, analyzes it, and recognizes what the speaker has said in the input voice data. For this recognition processing, any known method such as a general HMM (Hidden Markov Model) or a DP matching method may be used.

  The keyword determination unit 44 reads a keyword corresponding to the existing voice to be changed notified from the change receiving unit 41 from the keyword template, and inputs the input voice data in which the keyword is recognized and analyzed by the voice recognition unit 43. It is a means for detecting whether or not it is included in the sound by pattern matching. As a result of the pattern matching, when it is determined that the keyword of the corresponding existing voice is not included in the input voice, an error message is output to the display unit 10.

  When the keyword determination unit 44 determines that the keyword of the corresponding existing voice is included in the input voice, the voice registration unit 45 registers the input voice as the guidance voice in the voice guidance database instead of the corresponding existing voice. It is means to do. The existing voice data is stored in the voice guidance database as initial data in the voice by changing the guidance voice, and can always be returned to the initial state by a user instruction.

  The vehicle position detection, facility search, destination designation reception, map display, guide guidance, and the like are not the features of the present invention but may be the same as those in the prior art, and are therefore collectively shown as the navigation unit 40 and description thereof is omitted.

[2. Action)
The operation of the present embodiment configured as described above will be described with reference to the flowchart of FIG.
When the user selects the voice guidance change mode through the input unit 11 or the like (START), the change receiving unit 41 receives a voice guidance change request, so that the display unit 10 as shown in the example of the voice guidance change reception screen shown in FIG. A changeable voice list is displayed (S1), and selection of an existing voice to be changed is requested (S2). This process is repeated until the user selects an existing voice (NO in S2). When the user selects an existing voice to be changed (YES in S2), the existing voice data to be changed is specified. (S3), the fact is notified to the keyword determination unit 44.

  Next, the change receiving unit 41 confirms whether or not there is an input to the microphone 15 (S4). If there is an input (YES in S4), the signal processing unit 42 receives the analog audio signal and converts it into a digital signal (S5). On the other hand, if there is no input to the microphone 15 (NO in S4), the external storage device control unit 14 is then made to confirm whether there is an input from the external storage device 20 (S6). If there is an input from the external storage device 20 (YES in S6), the data is read (S7), and if there is no input (NO in S6), the communication device and the processing unit 13 are made to confirm the input from the portable terminal 21. (S8). Further, if there is an input from the portable terminal 21 (YES in S8), the input voice data is received (S9), and if there is no input (NO in S8), the process returns to S4 to confirm the input again. Perform (S4 to S9).

  The input voice data converted or read in S5, S7 or S9 is once recorded in the memory M, and the voice recognition unit 43 is notified that the conversion or reading has been performed. Then, the voice recognition unit 43 reads the input voice data from the memory M and performs voice recognition processing (S10).

  When the speech recognition unit 43 analyzes the input speech and determines what the speaker has said, the keyword determination unit 44 next selects a keyword corresponding to the existing speech to be changed notified from the change reception unit 41. It is read from the keyword template, and pattern matching is performed to determine whether or not the keyword is included in the input voice of the input voice data (S11).

  More specifically, if the existing voice is “left turn guidance” as shown in FIG. 3, the keyword template has the keyword of the existing voice number “2” as “tsugi” as shown in FIG. It is made up of “hidari” or “susetsu”, and it is determined whether or not these words are included in the input speech.

  For example, when the input voice is “next intersection to the right”, “tsugi” is included, but “hidari” and “susset” are not included, so the keyword condition is not satisfied (NO in S11). An error message is displayed on the display unit 10 (S12), and the user is requested to input voice again (S4).

  On the other hand, when the input voice is “next, left” (next, hiding), since “tsugi” and “hidari” are included, the voice registration unit 45 assumes that the keyword condition is satisfied (YES in S11). The input voice “next, left” is registered as an input voice in the voice guidance database (S13), and the input voice registration process is terminated (END).

  When the input voice is registered as described above, the guidance guidance in the navigation process in the navigation unit 40 is performed by the input voice registered in the voice guidance database.

[3. effect〕
According to the present embodiment as described above, an input voice input from a microphone, an external storage medium, or a portable terminal is analyzed by the voice recognition unit, and whether or not the keyword of the existing voice is included in the analyzed voice. By performing the determination in the keyword determination unit, it is possible to register an accurate input voice that matches the content of the existing voice.

  In addition, since the matching between the existing voice and the input voice is performed by pattern matching as to whether or not a predetermined keyword included in the existing voice is included in the input voice, it is allowed to match the contents of the existing voice and the input voice. By setting a keyword as a condition of the greatest common divisor, the degree of freedom of selection of the input voice is increased, and it is possible to register the input voice that is compatible with various preferences of the user and is guaranteed to be accurate.

  Furthermore, input voice input is not limited to a method using a microphone, and can be performed by various methods such as input from an external storage medium such as a flash memory card or USB memory, a portable terminal, and the like. Can be increased.

[4. Other embodiments]
In addition, this invention is not limited to the said embodiment, Other embodiments which are illustrated next are included. In the above embodiment, the voice guidance changing process in the setting stage of the device before the navigation process by the navigation unit has been described. The change process shown in FIG. 2 can be performed by accepting the change in accordance with the voice guidance change request (the process of the voice guidance switching unit).

  In such a case, based on the current position calculated by the navigation unit, the change receiving unit presents, for example, existing voice data to be changed at the next guidance point to the user as a selection target, It is also possible to determine the type of voice guidance at the next guidance point and to speak by obtaining the possibility of change.

  According to the other embodiments as described above, the user can change the existing voice at any time in response to the user's change request, such as during the navigation process, so that the user can provide voice guidance according to the driving situation at any time. , You will be able to select and register.

  In the present invention, the voice guidance can be changed not only from the existing voice to the input voice but also from the input voice to the existing voice. This change can also be made at any time by the change receiving unit. The input voice once registered is registered in the voice guidance database until another input voice is registered, and the change from the input voice to the existing voice and the change from the existing voice to the input voice can always be repeated.

The functional block diagram which shows the structure of embodiment of this invention. The flowchart which shows the process of embodiment of this invention. The figure which shows the example of the audio guidance change reception screen in embodiment of this invention. The figure which shows the example of the keyword template in embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Absolute position and direction detection part 2 ... Relative direction detection part 3 ... Vehicle speed detection part 4 ... Control part (Main CPU and its peripheral circuit)
5 ... ROM
6. Dynamic RAM (DRAM)
7. Static RAM (SRAM)
8 ... Video RAM (VRAM)
DESCRIPTION OF SYMBOLS 9 ... User interface part 10 ... Display part 11 ... Input part 12 ... Disk control part 13 ... Communication apparatus and processing part 14 ... External storage device control part 15 ... Microphone 20 ... External storage device 21 ... Portable terminal 40 ... Navigation part 41 ... Change receiving unit 42 ... signal processing unit 43 ... voice recognition unit 44 ... keyword determination unit 45 ... voice registration unit

Claims (7)

In a voice guidance device used for route guidance or the like in a navigation device, comprising a voice recognition unit for recognizing an input voice inputted from a predetermined input means, and a voice registration unit for registering the voice as a guidance voice.
A voice determination unit that determines whether the input voice corresponds to an existing voice that has already been registered;
A keyword is set for each existing voice in the existing voice,
The voice determination unit includes means for determining that the keyword is included in the input voice,
The voice registration device, wherein the voice registration unit registers the input voice when the keyword is included in the input voice. The existing and voice guidance by the input speech and voice guidance by voice, by operation of the user, the voice guidance device according to claim 1, characterized in that it comprises a voice guidance switching unit to enable always switched. 3. The voice guidance apparatus according to claim 1, wherein the input means includes an external storage means for storing a voice signal in addition to the sound collector. Voice guidance used for route guidance guidance and the like in a navigation device, in which a voice recognition process for recognizing an input voice inputted from a predetermined input means and a voice registration process for registering the voice as a guidance voice are executed by a control unit. In the device control method,
Including a voice determination process by which the control unit determines whether the input voice corresponds to an existing voice that has already been registered,
A keyword is set for each existing voice in the existing voice,
The voice determination process determines that the keyword is included in the input voice,
The method for controlling a voice guidance device, wherein the voice registration process is to register the input voice by the control unit when the keyword is included in the input voice. 5. The method of controlling a voice guidance device according to claim 4 , wherein voice guidance switching processing is performed so that the voice guidance based on the existing voice and the voice guidance based on the input voice can always be switched by a user operation. In a control program for a voice guidance device used for route guidance or the like in a navigation device, causing a control unit to recognize a voice input from a predetermined input unit and function as a voice registration unit that registers the voice as a guidance voice.
The control unit
Whether and corresponds with the existing voice the input speech has already been registered, is intended to function as a sound determination unit that Ru is determined,
A keyword is set for each existing voice in the existing voice,
The voice determination means determines that the keyword is included in the input voice,
The control program for a voice guidance device, wherein the voice registration means registers the input voice when the keyword is included in the input voice. 7. The voice guidance according to claim 6 , wherein the voice guidance based on the existing voice and the voice guidance based on the input voice cause the control unit to function as voice guidance switching means that can always be switched by a user operation. Device control program.

Images