JP7699471B2 - Voice guidance device, voice guidance method, and voice guidance program - Google Patents

Description

The present invention relates to a voice guidance device, a voice guidance method, and a voice guidance program.

Nowadays, for example, inside or outside a station or facility, voice guidance such as "The ticket gate is 5 meters ahead" or "The pedestrian crossing is 3 meters ahead" is provided to visually impaired users.

Patent Document 1 (JP Patent Publication 2020-125907A) discloses a voice guidance system for visually impaired people that provides voice guidance to a user (visually impaired person) passing through a station according to the direction of the user's movement. This makes it possible to avoid providing unnecessary voice guidance to a user passing through a station, thereby reducing the output of unnecessary voice guidance.

JP2020-125907A

However, with conventional voice guidance systems, including the voice guidance system for visually impaired people described in Patent Document 1, when multiple visually impaired users are located close to each other, it is difficult for them to know whether the voice guidance they are hearing is intended for them or for other visually impaired users.

For example, suppose multiple visually impaired users are located in the same place, and one user and the other user are walking in different directions. In this situation, suppose one user is given voice guidance such as "The ticket gate is five meters straight ahead." If the voice guidance given to one user is mistaken by the other user, who is walking in the opposite direction to the first user, as being intended for him/herself, the other user will be inconvenienced and will not be able to reach the ticket gate even if he/she walks five meters straight ahead.

The present invention was made in consideration of the above-mentioned problems, and aims to provide a voice guidance device, a voice guidance method, and a voice guidance program that enable visually impaired users to recognize that the voice guidance being output is intended for them, and to effectively provide voice guidance for each user without causing confusion.

In order to solve the above-mentioned problems and achieve the object, the present invention comprises a detection unit that detects visually impaired users and detects at least the current position of the visually impaired users by analyzing an image captured by a camera device, an assignment unit that generates guidance voices based on guidance voice data with different voice qualities assigned to each user when multiple visually impaired users are detected by the detection unit, and an output control unit that controls the output of the guidance voices generated based on the guidance voice data with different voice qualities assigned to each user via a voice output device corresponding to at least the current position of each visually impaired user detected by the detection unit.

According to the present invention, visually impaired users can be made aware that the voice guidance being output is intended for them. This allows voice guidance to function effectively for each user without causing confusion.

FIG. 1 is a diagram showing an example of a system configuration of a voice guidance system according to an embodiment of the present invention. FIG. 2 is a block diagram of an analysis device provided in the voice guidance system according to the embodiment. FIG. 3 is a diagram showing an example of a format of map data stored in the analysis device. FIG. 4 is a diagram showing an example of a format of guidance voice data stored in the analysis device. FIG. 5 is a schematic diagram of a user information table. FIG. 6 is a functional block diagram of each function realized by the CPU of the analysis device executing the voice guidance program. FIG. 7 is a flowchart showing the flow of the first half of the voice guidance operation of the voice guidance system according to the embodiment. FIG. 8 is a flowchart showing the flow of the latter half of the voice guidance operation of the voice guidance system according to the embodiment. FIG. 9 is a first schematic diagram for explaining a voice guidance operation in the voice guidance system according to the embodiment. FIG. 10 is a second schematic diagram for explaining the voice guidance operation in the voice guidance system according to the embodiment. FIG. 11 is a third schematic diagram for explaining the voice guidance operation in the voice guidance system according to the embodiment. FIG. 12 is a fourth schematic diagram for explaining the voice guidance operation in the voice guidance system according to the embodiment.

The following describes an embodiment of the voice guidance system according to the present invention with reference to the drawings.

(System Configuration)
Fig. 1 is a diagram showing a system configuration of a voice guidance system according to an embodiment. As shown in Fig. 1, the voice guidance system is configured by connecting a plurality of terminal devices 60 and an analysis device 3, which is a manager terminal device installed in, for example, a management room, via a wide area network such as the Internet or a private network such as a LAN (Local Area Network).

The terminal devices 60 are provided at different geographical locations, for example at predetermined intervals along a passageway through which users pass. Each terminal device 60 includes a camera device 1 and a speaker device 2. The camera device 1 is, for example, a fixed camera device, and captures images of users passing through a passageway or the like within a fixed imaging area. The camera device 1 may also be a camera device whose imaging area is changeable. The speaker device 2 is an example of an audio output device, and outputs audio guidance.

The analysis device 3 analyzes the images captured by the camera device 1 of each terminal device 60 to analyze the characteristics of visually impaired users. The analysis device 3 then assigns a different voice to each visually impaired user and outputs a guidance voice from the speaker device 2 provided in the position where the user moves. This makes it possible to provide voice guidance to each user without confusion, even if there are multiple visually impaired users in close proximity.

(Hardware configuration of the analysis device)
Fig. 2 is a block diagram showing a hardware configuration of the analysis device 3. As shown in Fig. 2, the analysis device 3 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, and a communication unit 14. The analysis device 3 also includes a HDD (Hard Disk Drive) 15, an input/output interface (input/output I/F) 16, and a communication interface (communication I/F) 17.

The communication unit 14 performs wired communication via a network such as the Internet or a LAN, as well as wireless communication such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). The HDD 15 stores a voice guidance program for providing voice guidance to visually impaired users, map data 50, guidance voice data 51, and a user information table 52.

As shown in FIG. 3, the map data 50 stores facility information and location information, including the names of facilities, tenants, ticket gates, elevator devices, etc., located in the geographic area where voice guidance is provided (hereinafter, the term "service area" is also used to mean the same thing), in association with output condition information indicating the conditions for providing voice guidance for the facilities.

The guidance voice data 51 is data for providing voice guidance based on the map data 50, and multiple audibly different guidance voice data 51 are stored. More specifically, the guidance voice data 51 is stored as divided voice data, which is voice data for each of various words, such as "I," "Taro," "Hanako," "5," "meters," "ahead," "to," "ticket gate," "exit," "is," "there," "black," "white," "color," "of," "cardigan," "for," "coat," "baseball," "hat," "long," "short," and "hair."

As an example, as shown in FIG. 4, a plurality of aurally different guidance voice data 51 are assigned a unique guidance voice ID, and are stored as male speaker guidance voice data 51 and female speaker guidance voice data 51. Even if the speakers are the same gender, the stored guidance voice data 51 is easy to distinguish aurally. Each male speaker guidance voice data 51 stores information indicating the voice frequency, sound pressure, pitch, and speech rate. Similarly, female speaker guidance voice data 51 stores guidance voice data 51 that is easy to distinguish aurally. Each female speaker guidance voice data 51 stores information indicating the voice frequency, sound pressure, pitch, and speech rate.

In the case of the voice guidance system of the embodiment, when there are multiple visually impaired users, factors such as the gender of the speaker, voice frequency, sound pressure, pitch, and speaking speed as shown in Figure 4 are used, and voice guidance is provided by assigning guidance voice data 51 that is easy for visually impaired users to distinguish.

Figure 5 shows a schematic diagram of the user information table 52. As shown in Figure 5, the user information table 52 is a table that stores the correspondence between visually impaired users and the guidance voice data 51 assigned to each user. As will be described in detail later, the table stores the correspondence between the user ID, information on the characteristics (personality characteristics) of the user, and the ID (guidance voice ID) of the guidance voice data 51 assigned to the user.

The input/output I/F 16 is connected to a display unit 18 and an operation unit 19 when necessary. The communication I/F 17 is connected to the network 5 via a network cable when necessary.

(Functional configuration of the analysis device)
6 is a functional block diagram of each function realized by software when the CPU 11 executes the voice guidance program stored in the HDD 15. As shown in this Fig. 6, the CPU 11 executes the voice guidance program to function as a video acquisition unit 21, a map data acquisition unit 22, an image analysis unit 23, an output voice allocation unit 24, a communication control unit 25, a speaker switching unit 26, and an emergency processing unit 27.

The video acquisition unit 21 acquires captured images of users traveling within the geographical area captured by each camera device 1. The map data acquisition unit 22 acquires map data 50 corresponding to the longitude and latitude of the geographical area captured by each camera device 1 from the HDD 15. The image analysis unit 23 is an example of a detection unit, and analyzes the personal characteristics of the user based on the captured images captured by each camera device 1, and also determines whether or not the user has a visual impairment, etc.

When it is determined that the user is visually impaired, the image analysis unit 23 determines whether or not information on characteristics (personal characteristics) that matches the analyzed information on the user's characteristics (personal characteristics) is stored in the user information table 52. If it is not stored in the user information table 52, the image analysis unit 23 issues a user ID that uniquely represents the analyzed user, and stores the issued user ID in the user information table 52 in association with the analyzed information on the user's characteristics (personal characteristics).

The image analysis unit 23 also detects the current position of the user based on the longitude and latitude corresponding to each coordinate of the captured image captured by the camera device 1. Furthermore, the image analysis unit 23 detects the direction of movement of the user from the difference in the current positions of the same user captured in a series of captured images over several frames, for example. The image analysis unit 23 also detects the presence or absence of a "movement calling for help" by a visually impaired user, such as raising a white cane about 50 cm above the head or waving the white cane from side to side in front of the user's face.

The output voice allocation unit 24 is an example of an allocation unit, and allocates guidance voice data 51 stored in the HDD 15 to a visually impaired user detected by the image analysis unit 23. Specifically, when a new user is registered by the image analysis unit 23, the output voice allocation unit 24 allocates another guidance voice data 51 with a voice quality different from the guidance voice already allocated from the guidance voice data 51 stored in the HDD 15. Then, the output voice allocation unit 24 stores the ID (guidance voice ID) of the guidance voice data 51 in the user information table 52 as the guidance voice for the new user. Furthermore, the output voice allocation unit 24 refers to the user information table 52, and generates a guidance voice for each user based on the guidance voice data 51 with a different voice quality allocated to each user.

The speaker switching unit 26 is an example of an output control unit, and controls switching of the speaker device 2 that outputs the guidance voice according to the current location of the visually impaired user, and outputs the guidance voice generated by the guidance voice data 51 of the voice quality assigned to the user. As a result, the voice guidance is performed following the movement of each user with the guidance voice of the voice quality assigned to each user. The communication control unit 25 communicates with each terminal device 60, and acquires the captured image captured by the camera device 1 and transmits the guidance voice to the speaker device 2. When the image analysis unit 23 analyzes the action of the visually impaired user, etc., to request help, the emergency processing unit 27 issues an emergency notification to the administrator, etc. based on the analysis result. The emergency processing unit 27 also controls the output of a message that a staff member will immediately come to the rescue via the speaker device 2 corresponding to the location of the user requesting help.

In this example, the video acquisition unit 21 to the emergency processing unit 27 are implemented as software using a voice guidance program. However, all or part of these may be implemented as hardware such as an integrated circuit (IC).

The voice guidance program may be provided by recording it in the form of file information in an installable or executable format on a recording medium that can be read by a computer device, such as a CD-ROM or a flexible disk (FD). The voice guidance program may be provided by recording it on a recording medium that can be read by a computer device, such as a CD-R, a DVD (Digital Versatile Disc), a Blu-ray (registered trademark) disc, or a semiconductor memory. The voice guidance program may be provided by installing it via a network such as the Internet. The voice guidance program may be provided by being pre-installed in a ROM or the like within the device.

(Voice guidance operation)
7 and 8 are flowcharts showing the flow of the voice guidance operation in the voice guidance system of the embodiment. Of these, Fig. 7 is a flowchart showing the flow of the first half of the voice guidance operation, and Fig. 8 is a flowchart showing the flow of the second half of the voice guidance operation.

(Step S1)
First, in step S1 of the flowchart in FIG. 7, the video acquisition unit 21 acquires images of users (passersby) captured by each camera device 1.

(Step S2)
After step S1, in step S2, the image analysis unit 23 analyzes the user's features (personal features), current location, and moving direction from the captured image acquired in step S1.

As for the user's characteristics, the image analysis unit 23 detects the user's age and gender using a predetermined algorithm. In addition, the image analysis unit 23 detects the user's characteristics such as clothing, clothing color, belongings such as a handbag or backpack, and the colors of the belongings by analyzing the captured image. These characteristics are called personal characteristics.

The image analysis unit 23 also detects the current position of the user based on the longitude and latitude corresponding to each coordinate of the captured image captured by the camera device 1. The image analysis unit 23 also detects the direction of movement of the user from the difference in the current positions of the same user captured in a series of captured images, for example, several frames.

(Step S3)
After step S2, in step S3, the image analysis unit 23 determines whether the user shown in the captured image is a normal person or a visually impaired user.

As one example, a visually impaired user owns a blind safety cane, which is a white cane. In contrast, an elderly person who is not visually impaired but has difficulty walking uses a brown or black cane, etc. For this reason, the image analysis unit 23 determines whether or not a user in a captured image is visually impaired based on whether or not the cane owned by the user is white.

In addition, visually impaired users have a unique movement of lightly tapping the ground with a blind safety stick as they walk to check for the presence or absence of obstacles. The image analysis unit 23 uses the presence or absence of such a unique movement as a factor for determining whether or not a user is visually impaired.

In addition, visually impaired users may be accompanied by guide dogs. Ordinary dogs are fitted with a collar or harness with a single string-like lead attached. Guide dogs, on the other hand, are fitted with a uniquely shaped harness called a "U-shaped harness" or "bar-handle harness." The image analysis unit 23 also uses the shape of such a harness as a factor for determining whether or not the user is visually impaired.

Also, "U-shaped harnesses" or "bar-handle harnesses" are often white in color. For this reason, the image analysis unit 23 also uses the color of the harness worn by the dog as a factor for determining whether the user with the dog is visually impaired.

Furthermore, if a user enters a place where entry with dogs is restricted, the dog is likely to be a guide dog and the user is likely to be visually impaired. For this reason, the image analysis unit 23 determines that a user who is bringing a dog to a place where entry with dogs is restricted is a visually impaired user.

In addition, guide dogs often walk on the left side of the user's direction of travel. For this reason, the image analysis unit 23 uses whether or not a user is accompanied by a dog that walks on the left side of the user's direction of travel as a factor for determining whether or not the user is visually impaired.

In addition, the image analysis unit 23 also uses factors such as whether or not the user is wearing sunglasses, whether or not there is a handicapped mark, etc., to determine whether or not the user is visually impaired.

(Step S3: No → Go to step S1)
Next, if the users appearing in the captured image are all able-bodied individuals (step S3: No), the process returns to step S1, and the processes of steps S1 to S3 are repeatedly performed by the image analysis unit 23. On the other hand, if it is determined that the users appearing in the captured image are visually impaired users (step S3: Yes), the process proceeds to step S4.

(Step S4)
In step S4, the image analysis unit 23 refers to the user information table 52 based on the characteristics (personality characteristics) of the user determined to be visually impaired, obtained by analyzing the captured image in step S2, and determines whether the user is already a registered user based on whether information of the same characteristics (personality characteristics) is stored in correspondence with the user ID and guidance voice ID.

(Step S4: No → Go to step S14)
If the user is not a known user registered in the user information table 52 (step S4: No), the process proceeds to step S14.

(Step S14: Newly assign a guidance voice ID)
In step S14, the image analysis unit 23 issues a new user ID to the user determined in step S4 to be not a known user. The image analysis unit 23 associates the issued user ID with the analyzed information on the user's characteristics (personal characteristics) and stores them in the user information table 52. In addition, the output voice allocation unit 24 assigns guidance voice data 51 of a voice quality that is not currently assigned to another user. Then, the output voice allocation unit 24 associates the ID (guidance voice ID) of the guidance voice data 51 with the user ID and characteristics (personal characteristics) of the user and stores them in the user information table 52. That is, the output voice allocation unit 24 assigns guidance voice data 51 with a difference in auditory quality as the guidance voice data 51 for each user.

(Detailed Description of Step S14)
The auditory difference is generated by making one or more of the following different: gender, voice frequency (e.g., formant frequency), sound pressure, pitch, speaking speed, etc. Specifically, for example, when there are two visually impaired users, the output voice allocation unit 24 allocates guidance voice data 51 of a male voice to one user, and voice data of a female voice to the other user. Alternatively, the output voice allocation unit 24 allocates guidance voice data 51 of a male voice with a high voice frequency and a fast speaking speed to one user, and allocates guidance voice data 51 of a male voice with a low voice frequency and a slow speaking speed to the other user, even if both users are male.

In this way, by assigning guidance voice data 51 with different voice qualities to each user based on gender, voice frequency, sound pressure, pitch, and speaking speed, it is possible to make it easier for each user to recognize in advance the voice quality of the guidance voice that will be used for them.

(Step S14 → Step S15: Generation of Guidance Voice (for Pre-recognition))
After step S14, in step S15, the output voice allocation unit 24 generates a voice guidance including the "pre-recognition voice guidance" described later, using the voice guidance data 51 corresponding to the voice guidance ID assigned in step S13. At this time, the output voice allocation unit 24 generates a voice guidance including a proper noun (e.g., "Taro" or "Hanako") indicating the speaker of the voice guidance data 51, which is associated with each voice guidance data 51. By outputting the voice guidance generated in this manner to the user, the user can be made aware in advance that voice guidance will be provided to him or her by a voice guidance from a specific speaker.

(Detailed Description of Step S15)
The output voice allocation unit 24 generates a voice guidance including a proper noun indicating the speaker who will provide the voice guidance, such as "Taro" or "Hanako", for example, "I, Hanako, will guide you", using the voice guidance data 51 corresponding to the assigned voice guidance ID. This allows the user to recognize in advance that the voice guidance for him/her will be provided in a voice with the voice quality of, for example, "Hanako".

Note that this example is an example of adding a proper noun "given name" such as "Taro" or "Hanako." In addition, a proper noun "surname" or "name" may be added, or other proper nouns such as place names, country names, building names, etc. may be added.

After step S15, the process proceeds to step S6.

(Step S4: Yes)
On the other hand, in step S4, if the user is a known user who has already been registered in the user information table 52 (step S4: Yes), the process proceeds to step S5.

(Step S5: Acquire assigned voice guidance ID)
In step S5, the output voice allocation unit 24 obtains the guidance voice ID assigned to the user from the user information table 52. Then, the process proceeds to step S6.

In step S6, the output voice allocation unit 24 generates a guidance voice including voice guidance of facilities corresponding to the user's current location and direction of travel using the guidance voice data 51 corresponding to the guidance voice ID acquired in step S5 or assigned in step S13.

(Detailed Description of Step S6)
Specifically, when the user's current location is, for example, near a store, the output voice allocation unit 24 reads out guidance voice data 51 corresponding to the assigned guidance voice ID for each of various words such as "on your right", "to", "store", "A", "there", "there is", etc. from the HDD 15. In addition, the output voice allocation unit 24 combines the read guidance voice data 51 to generate a guidance voice having voice guidance content corresponding to the user's current location and moving direction, such as "Store A is on your right".

(Steps S7 to S8: Determination of speaker device, output of guidance voice)
After step S6, the process proceeds to step S7, where the speaker switching unit 26 determines the speaker device 2 that outputs the guidance voice based on the current location of each user, or the current location and moving direction. Then, the process proceeds to step S8, where the speaker switching unit 26 controls the output of the guidance voice generated by the output voice allocation unit 24 in step S6, which includes the voice guidance of facilities, etc. corresponding to the user's current location and moving direction, via the speaker device 2 determined in step S7. Alternatively, the speaker switching unit 26 controls the output of the pre-recognition voice guidance generated by the output voice allocation unit 24 in step S14, and the guidance voice including the voice guidance of facilities, etc. corresponding to the user's current location and moving direction, via the speaker device 2 determined in step S7.

(Specific example of steps S15 to S8)
Here, a flow of generating a guidance voice including the "pre-recognition voice guidance" and outputting it from the speaker device 2 will be described with a concrete example. For example, if the user's characteristics analyzed by the image analysis unit 23 are a woman wearing a black coat, the output voice allocation unit 24 reads out guidance voice data 51 corresponding to the guidance voice ID assigned to the user for each of various words such as "black,""color,""of,""coat,""wear,""woman,""of," and "person" from the HDD 15. In addition, the output voice allocation unit 24 combines the read guidance voice data 51 to generate guidance voice including voice guidance (pre-recognition voice guidance) for the user to recognize that the voice guidance is for the analyzed user, such as "a woman wearing a black coat," and voice guidance based on the user's current position and moving direction.

Then, the speaker switching unit 26 determines the speaker device that will output the guidance voice based on the current position and moving direction of the user, for example, to be the speaker device 2 of the terminal device 60 in which the camera device 1 that captured the analyzed captured image of the user is provided. The speaker switching unit 26 controls the output of the above-mentioned guidance voice via the determined speaker device 2. This allows the user to hear the above-mentioned pre-recognition voice guidance, so that the user can be made aware in advance that the voice guidance that is about to be output is voice guidance for him/her and the voice quality of the guidance voice.

When voice guidance for one or more users is started in this manner, the speaker switching unit 26 selects the speaker device 2 corresponding to the user's current position and direction of movement, and outputs the guidance voice in the voice quality assigned to that user. In this manner, the voice guidance for the user is always provided in the guidance voice with the initially assigned voice quality. Therefore, even if there are multiple visually impaired users in close proximity, the voice guidance is provided to each user in a different voice quality, so that each visually impaired user can always recognize that the voice guidance being output is for them, and will not be confused. Therefore, the voice guidance can be provided effectively for each user.

(Steps S9 and S10)
Next, after step S8, the process proceeds to step S9 in the flowchart of Fig. 8, where the output voice allocation unit 24 determines whether the user has moved outside the service area. Specifically, the output voice allocation unit 24 refers to the last access date and time to the user information table 52, and if it is more than a certain time (e.g., one hour) before the current time, it determines that the user has moved outside the service area. If it is determined that the user has moved outside the service area (step S9: Yes), the output voice allocation unit 24 erases information about the user from the user information table 52 (step S10).

Alternatively, the image analysis unit 23 determines whether the user has moved outside the facility by analyzing camera footage of the facility entrance or exit (step S9). If the camera footage confirms that the user has moved outside the facility (step S9: Yes), the output audio allocation unit 24 deletes information about the user from the user information table 52 (step S10).

(Specific example of voice guidance for multiple users)
More specifically, Figs. 9 to 12 are diagrams that show voice guidance provided to visually impaired users A and B. First, as shown in Fig. 9, assume that user A goes straight from the left and user B goes straight from the right in a first aisle of a store. A second aisle is provided to the first aisle so as to form a so-called T-junction. Along the first and second aisles, terminal devices 60a to 60h corresponding to the terminal device 60 shown in Fig. 1 are placed at predetermined intervals. User A and user B are photographed by the camera device 1 of terminal device 60a and the camera device 1 of terminal device 60f, respectively, and it is analyzed that user A has the characteristics of "a woman wearing a black coat" and user B has the characteristics of "a man wearing a gray suit."

User A is walking near the speaker device 2 of terminal device 60a in the first passage, and user B is walking near the speaker device 2 of terminal device 60f in the first passage. In this case, the speaker switching unit 26 selects the speaker device 2 of terminal device 60a as the speaker device that outputs audio guidance to user A, and selects the speaker device 2 of terminal device 60f as the speaker device that outputs audio guidance to user B. In addition, the output audio allocation unit 24 assigns guidance audio data 51 (guidance audio ID: M1) of male speaker Taro to user A, and assigns guidance audio data 51 (guidance audio ID: F1) of female speaker Hanako, who has a different voice quality from that of male speaker Taro, to user B.

The speaker switching unit 26 outputs, via the speaker device 2 of the terminal device 60a, guidance voice including pre-recognition voice guidance, for example, "Lady in a black coat, I, Taro, will guide you," generated from the guidance voice data 51 with the guidance voice ID M1 assigned to user A. This allows user A to recognize that the voice guidance for him/her will be provided by the voice of a male person named Taro. Note that by providing pre-recognition voice guidance based on personal characteristics as described above, user A can be made to further recognize that the upcoming voice guidance is for him/her.

Similarly, the speaker switching unit 26 outputs, via the speaker device 2 of the terminal device 60f, guidance voice including pre-recognition voice guidance, for example, "Male in a gray suit, I am Hanako and will guide you. Store B is ahead on the left," generated from guidance voice data 51 with guidance voice ID F1 assigned to user B. This allows user B to recognize that the voice guidance for him/her will be provided by the voice of a female person, Hanako. Note that by providing pre-recognition voice guidance based on personal characteristics as described above, user B can be made to further recognize that the upcoming voice guidance is for him/her.

Next, as shown in FIG. 10, assume that user A and user B, who are moving straight ahead, advance to a position close to the second passage. In this case, the speaker switching unit 26 selects the speaker device 2 of the terminal device 60b as the speaker device 2 that outputs the guidance voice for user A, and selects the speaker device 2 of the terminal device 60d as the speaker device 2 that outputs the guidance voice for user B.

The speaker switching unit 26 then outputs, via the speaker device 2 of the terminal device 60b, a guidance voice generated from the guidance voice data 51 with a guidance voice ID of M1 assigned to user A, for example, "Ahead is a T-junction. Turn right to store A, and go straight to store B." The speaker switching unit 26 also outputs, via the speaker device 2 of the terminal device 60d, a guidance voice generated from the guidance voice data 51 with a guidance voice ID of F1 assigned to user B, for example, "Ahead is a T-junction. Turn left to store A."

Next, as shown in FIG. 11, assume that user A and user B approach a T-junction at almost the same time. In this case, the speaker device selected is the speaker device 2 of the same terminal device 60c. The speaker switching unit 26 outputs, via the speaker device 2 of the terminal device 60c, a guidance voice generated from the guidance voice data 51 with the guidance voice ID M1 assigned to user A, for example, "This is a T-junction. Turn right to store A, and go straight to store B." The speaker switching unit 26 also outputs, via the speaker device 2 of the terminal device 60c, a guidance voice generated from the guidance voice data 51 with the guidance voice ID F1 assigned to user B, for example, "This is a T-junction. Turn left to store A."

In the example of FIG. 11, users A and B are located close to each other, but each user A and B recognizes the voice quality of the guidance voice for them in advance. In addition, the guidance voice with the guidance voice ID M1 used in the voice guidance for user A and the guidance voice with the guidance voice ID F1 used in the voice guidance for user B have different voice qualities, so users A and B can distinguish the guidance voice for themselves from the guidance voice for the other user without confusing them. As a result, even if the voice guidance for users A and B is output almost simultaneously through the same speaker device 2, users A and B can distinguish the guidance voices with different voice qualities and act according to the voice guidance for themselves. This allows the voice guidance for users A and B to function effectively.

Furthermore, as shown in FIG. 12, when user A moves straight down the first passage and reaches a position close to the speaker device 2 of terminal device 60e, the speaker switching unit 26 outputs, via the speaker device 2 of terminal device 60f, a guidance voice generated from the guidance voice data 51 with guidance voice ID M1 assigned to user A, for example, "You will soon arrive at store B. Store B is on your right." This allows user A to recognize that he or she has moved to store B.

When user B enters the second passage and moves to a position close to the speaker device 2 of the terminal device 60h, the guidance voice generated by the guidance voice data 51 of the guidance voice ID F1 assigned to user B, for example, "You will soon arrive at store A. Store A is on your left." is output via the speaker device 2 of the terminal device 60h. This allows user B to recognize that he or she has moved close to store A.

In this way, by switching the speaker device 2 in response to the user's movement and outputting guidance voices with different voice qualities assigned to each user, it is possible to provide voice guidance to each user without causing confusion.

(Emergency Processing)
8, the image analysis unit 23 detects whether or not the visually impaired user is making a "movement calling for help," such as, for example, raising the white cane about 50 cm above the head or waving the white cane from side to side in front of the user's face. If such a "movement calling for help" is not detected (step S11: No), the process returns to step S1.

In contrast, if a "movement calling for help" is detected (step S11: Yes), the emergency processing unit 27 issues an emergency notification, for example via the display unit 18, indicating that the visually impaired user is calling for help (step S12).

In addition, in step S13, the speaker switching unit 26 issues voice guidance such as "An emergency notification has been sent to the administrator. Help will be on the way soon, so please wait for a while" via the speaker device 2 corresponding to the current location of the user requesting help. That is, the speaker switching unit 26 issues voice guidance informing the user that the administrator has been contacted in response to the request for help, and asking the user to wait for a while, in a guidance voice with a voice quality assigned to the user. This allows the visually impaired user who has requested help to recognize that the administrator or the like is acting in response to his or her request for help, providing a sense of security. In addition, when this emergency notification is received, an assistant such as an administrator or security guard can respond by going directly to the location of the user requesting help to provide assistance.

(Effects of the embodiment)
As is clear from the above description, when multiple visually impaired users are located close to each other, the voice guidance system of the embodiment generates a guidance voice by assigning guidance voice data 51 with a different voice quality to each user. Then, the guidance voice with the assigned voice quality is output through the speaker device 2 corresponding to the moving position of each user. As a result, even when multiple visually impaired users are located in the same place, each user can easily distinguish the guidance voice for himself/herself, and the voice guidance can be effectively performed for each user.

In addition, by analyzing personal characteristics and providing voice guidance (pre-recognition voice guidance) in an assigned voice quality that helps the user recognize that the voice is directed at them, it becomes easier for each user to distinguish the voice guidance intended for them from others.

In addition, by including a proper noun indicating the speaker providing the audio guidance, such as "Taro" or "Hanako," in the audio guidance, each user can be made more aware of the audio guidance that is directed at them.

In addition, the speaker device 2 that outputs the guidance voice is switched according to the user's movement, so that the same speaker device 2 is always outputting the guidance voice, which can prevent the inconvenience of the voice guidance becoming a noise for able-bodied people, staff at nearby stores, nearby residents, etc.

In the above-described embodiment, each visually impaired user (passerby) is uniquely identified by using a user information table 52 in which the personal characteristics of each visually impaired user (passerby) are registered. However, this is not limiting, and the following may also be used.

For example, a visually impaired user (passerby) is provided with a wireless tag such as a BLE tag that transmits radio waves including the user's own identification information. BLE is an abbreviation for "Bluetooth (registered trademark) Low Energy." In addition, a receiving device for the radio waves including the user's own identification information transmitted by the wireless tag is provided in the terminal device 60 together with the camera device 1 and the speaker device 2, for example.

The receiving device receives radio waves from the wireless tag and transmits the identification information contained in the radio waves to the analysis device 3 via the network 5. The analysis device 3, together with the receiving device that received the identification information, analyzes the captured image captured by the camera device 1 provided in the terminal device 60, detects an image of the user, associates the received identification information with the detected image, and registers the image in a database. This allows each visually impaired user (passerby) to be uniquely identified, as described above.

Finally, the above-described embodiment is presented as an example and is not intended to limit the scope of the present invention. This novel embodiment can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. Furthermore, the embodiment and modifications of the embodiment are included in the scope and gist of the invention, and are included in the scope of the invention and its equivalents described in the claims.

Reference Signs List 1 Camera device 2 Speaker device 3 Analysis device 5 Network 11 CPU
12 ROM
13 RAM
14 Communication unit 15 HDD
16 Input/Output Interface (Input/Output I/F)
17 Communication interface (communication I/F)
18 Display unit 19 Operation unit 21 Video acquisition unit 22 Map data acquisition unit 23 Image analysis unit 24 Output audio allocation unit 25 Communication control unit 26 Speaker switching unit 27 Emergency processing unit 50 Map data 51 Guidance audio data 52 User information table

Claims (6)

a detection unit that detects a visually impaired user and detects at least a current location of the visually impaired user by analyzing an image captured by the camera device;
an allocation unit that generates a guidance voice based on guidance voice data of different voice qualities assigned to each user when a plurality of visually impaired users are detected by the detection unit;
an output control unit that controls output of the guidance voice generated based on the guidance voice data of different voice qualities assigned to each user via a voice output device corresponding to at least a current position of each visually impaired user detected by the detection unit;
A voice guidance device comprising: The voice guidance device according to claim 1, wherein the allocation unit allocates guidance voice data that differs in at least one or more of gender, voice frequency, sound pressure, pitch, and speaking speed to each visually impaired user. The detection unit detects characteristics of each visually impaired user,
The voice guidance device according to claim 1 or 2, wherein the output control unit performs pre-recognition voice guidance indicating the characteristics of each detected user using guidance voice data of a voice quality assigned to each user. The voice guidance device according to claim 1 , wherein the allocation unit generates the guidance voice data including a proper noun indicating a speaker providing the voice guidance when generating the guidance voice data. a detection step in which a detection unit detects a visually impaired user by analyzing an image captured by the camera device, and detects at least a current position of the visually impaired user;
an allocation step in which, when a plurality of visually impaired users are detected in the detection step, an allocation unit generates a guidance voice based on guidance voice data of different voice qualities assigned to each user;
an output control step in which an output control unit controls output of the guidance voice generated based on the guidance voice data of different voice qualities assigned to each user via a voice output device corresponding to at least a current position of each visually impaired user detected in the detection step;
The voice guidance method includes: Computer,
a detection unit that detects a visually impaired user and detects at least a current location of the visually impaired user by analyzing an image captured by the camera device;
an allocation unit that generates a guidance voice based on guidance voice data of different voice qualities assigned to each user when a plurality of visually impaired users are detected by the detection unit;
functioning as an output control unit that controls output of the guidance voice generated based on guidance voice data of different voice qualities assigned to each user via a voice output device corresponding to at least the current position of each visually impaired user detected by the detection unit;
A voice guidance program that features:

Images