JP5807451B2 - Voice processing device, voice processing method, program, and guidance system - Google Patents

Description

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

  In recent years, guidance and guidance techniques for guiding a user to a target position have become widespread. For example, a car navigation system, which is a typical example of a guidance and guidance device, creates a route from a current position to a target position, and guides the user to the target position by a combination of map display and voice output based on the route.

  On the other hand, it is difficult for a visually handicapped person to guide to a target position depending on the map display. For this reason, a technique for guiding a user who is visually impaired by only sound output has been proposed.

  For example, Patent Document 1 below proposes a guidance and guidance device that detects the direction of a user, determines a direction in which the user should proceed based on the direction of the user, and notifies the direction to proceed by voice. .

JP 2002-257581 A

  However, in the guidance and guidance device of Patent Document 1, the user cannot intuitively grasp the correct traveling direction and how to return when the user leaves the route. For example, in order to know whether the user is moving in the right direction, the user must confirm with the guidance and guidance device each time by uttering a voice toward a microphone. In addition, when the user is away from the route, the user cannot easily grasp the route and the start position of the movement.

  Therefore, the present invention provides a new and improved voice processing device, voice processing method, program, and guidance system that make it possible to intuitively grasp the correct direction of travel and how to return when leaving a route. It is something to try.

According to the present invention, a detection unit that detects a user position, a first virtual sound source position corresponding to a start position that is the user position detected by the detection unit when guidance to the user is started, and the above a sound source setting unit that sets a second virtual sound source position corresponding to the position existing on the path between the target position or the user position and the target position of the induction, the detected the user position by the detection unit The first audio data in which the first virtual sound source position is perceived by the user as the sound source position and the second sound data in which the second virtual sound source position is perceived by the user as the sound source are used. and a creating unit, the first sound data and the second audio data has a different voice pattern, the audio creation unit, the first audio data and the second audio data It places at different positions on the time axis, the sound processing apparatus is provided.

  The first voice data and the second voice data have different voice patterns, and the voice creation unit arranges the first voice data and the second voice data at different positions on the time axis. May be.

  The detection unit further detects the orientation of the user, and the voice creation unit detects each of the first voice data or the second voice data as the user position or the user orientation, and the first voice data. May be created based on a relative relationship with each of the virtual sound source positions or the second virtual sound source positions.

  The voice creation unit may create each of the first voice data and the second voice data with a volume or a voice pattern corresponding to the relative relationship.

  The voice creation unit may arrange each of the first voice data and the second voice data at different positions on the time axis at a frequency according to the relative relationship.

  The relative relationship is the distance between the user position and the first virtual sound source position or the second virtual sound source position, or the direction of the user and the direction of the first virtual sound source position or the first virtual sound source position. An angle formed by the direction of the two virtual sound source positions may be included.

  The voice creation unit may create the first voice data and the second voice data in a stereo format.

  The voice creation unit may create the first voice data and the second voice data by convolution of a head-related transfer function.

According to the present invention, the detection unit detects the user position, and the sound source setting unit detects the user position according to the start position that is the user position detected by the detection unit when the guidance to the user is started. 1 of the virtual sound source position, and a step of setting a second virtual sound source position corresponding to the position existing on the path between the target position or the user position and the target position of the induction, sound creation section above First audio data in which the first virtual sound source position set by the sound source setting unit is perceived by the user as a sound source position using the user position detected by the detection unit , and the first sound data and creating a different an audio pattern, the second sound data set the second virtual sound source position by the sound source setting unit is perceived by the user as sound source and, Serial sound creation unit and placing at different positions on the time axis of the first audio data and the second audio data, the audio processing method is provided.

According to the present invention, the computer includes a detection unit that detects a user position, and a first virtual corresponding to the start position that is the user position detected by the detection unit when guidance to the user is started. sound source location, and a sound source setting unit that sets a second virtual sound source position corresponding to the position existing on the path between the target position or the user position and the target position of the induction, is detected by the detection unit In addition, using the user position, the first virtual sound source position is perceived by the user as a sound source position, and the second virtual sound source has a sound pattern different from the first sound data. create a second audio data source position is perceived to the user as a sound source, a sound creation unit for disposing the first voice data and the second audio data at different positions on the time axis Program for functioning as is provided.

According to the present invention, there is provided a guidance system including a sensor, a voice output device, and a voice processing device, wherein the voice processing device includes a detection unit that detects a user position based on an input from the sensor, and a user A first virtual sound source position corresponding to the start position which is the user position detected by the detection unit at the time when the guidance for the guidance is started , and a route between the guidance target position or the user position and the target position a sound source setting unit that sets a second virtual sound source position corresponding to the position that is present above using the detected the user position by the detection unit, the user of the first virtual sound source position as a sound source position It comprises first audio data to be perceived, and the second virtual sound source position and the voice generating unit configured to generate a second audio data that is perceived to the user as a sound source, the first speech Over data and the second audio data has a different voice pattern, the audio creation unit arranges the first audio data and the second audio data at different positions on the time axis, the sound The output device is provided with a guidance guidance system that outputs the voice of the first voice data and the voice of the second voice data.

  As described above, according to the voice processing device, the voice processing method, the program, and the guidance and guidance system according to the present invention, it is possible to intuitively grasp the correct traveling direction and how to return when leaving the route.

It is explanatory drawing which shows an example of the schematic structure of the guidance system which concerns on one Embodiment. It is a block diagram which shows an example of a structure of the audio | voice processing apparatus which concerns on one Embodiment. It is explanatory drawing for demonstrating the example of the direction of the user detected. It is explanatory drawing for demonstrating the example of the route produced by the guidance information creation part. It is explanatory drawing for demonstrating preparation of the audio | voice data by an audio | voice preparation part. It is explanatory drawing for demonstrating preparation of the audio | voice data until a user turns to the direction of a target position. It is explanatory drawing for demonstrating creation of the audio | voice data at the time of a user moving linearly toward a target position. It is explanatory drawing for demonstrating creation of the audio | voice data when a user changes direction in the middle of a movement. It is explanatory drawing for demonstrating arrangement | positioning on the time-axis of the audio | voice data by an audio | voice preparation part. It is explanatory drawing for demonstrating the modification of the setting of a virtual sound source position. It is a flowchart which shows an example of the schematic flow of the audio | voice process which concerns on one Embodiment.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

  In the following, <1. Introduction>, <2. Schematic configuration of guidance system>, <3. Configuration of speech processing apparatus>, <4. Embodiments of the present invention will be described in the order of processing flow>.

<1. Introduction>
In recent years, guidance and guidance techniques for guiding a user to a target position have become widespread. For example, a car navigation system, which is a typical example of a guidance and guidance device, first acquires current position information from a GPS (Global Positioning System). Then, the car navigation system creates a route from the current position to the target position, and guides the user to the target position by a combination of map display and audio output based on the route. According to the car navigation system, the user can browse the displayed map to determine the positional relationship between the user's position and the target position or the start position of the guidance, whether or not the vehicle is moving in the correct direction, and away from the route. It is possible to easily grasp whether or not there is. In addition, the user can obtain useful information for reaching the target position by listening to the sound for notifying the timing of turning right or left. As described above, many guide guidance techniques guide a user to a target position by a combination of map display and voice.

  On the other hand, since it is difficult for a visually handicapped person to view a displayed map, it is difficult to guide a user who is a visually handicapped person to a target position depending on the map display as described above. For this reason, a technique for guiding a user who is visually impaired by only sound output has been proposed. As an example, there has been proposed a guidance and guidance device that detects a user's direction, determines a direction in which the user should travel based on the user's direction, and notifies the direction to travel by voice (Japanese Patent Laid-Open No. 2002-2002). No. 2,57581).

  However, with the conventional guidance and guidance technology, the user cannot intuitively grasp the correct traveling direction and how to return when the user leaves the route. For example, when using the above-described guidance and guidance device as an example, in order to know whether the user is moving in the right direction, the user must confirm with the guidance and guidance device each time by uttering a voice toward a microphone. . Further, when the user has left the route, the user cannot easily grasp that the user has left the route and how to return to the original route or the start position of the movement.

  The inventor has recognized the above problems of the guidance and guidance technology, and has studied the guidance and guidance technology that makes it possible to intuitively grasp the correct traveling direction and how to return when leaving the route. And in the research of the guidance technology, the inventor paid attention to the “heterogeneous squealing method”.

  The heterogeneous squealing method is a voice guidance method that is introduced to traffic lights so that visually impaired people can safely cross the road ("No. 77 of the Tokyo Metropolitan Police Department, No. 77, visually impaired person additional device for visually impaired persons, October 22, 2003." (See “Establishment of Installation / Operation Guidelines”). According to the heterogeneous squealing method, a traffic light located at one pedestrian crossing emits a sound such as “cuckoo” or “piyo”, and a traffic light located at the other pedestrian crossing ends is “cuckoo” or “piyopiyo”. And so on. The user can intuitively grasp where the two pedestrian crossings exist by listening to these sounds. In addition, since these voices heard by the user change in accordance with the movement of the user, the user can intuitively know from the change in the voice whether he / she is moving in the correct direction of travel or not away from the pedestrian crossing. Can be grasped. In addition, since the sound is emitted from the traffic lights at both ends of the pedestrian crossing, the user can return to the pedestrian crossing or to the end of the pedestrian crossing where he / she was. As described above, the heterogeneous squealing method has an advantage that it is possible to intuitively grasp the correct traveling direction and how to return when leaving a pedestrian crossing. Since the two traffic lights emit different sounds, the user can distinguish between the two traffic lights (or two pedestrian crossing ends). Moreover, since these voices are emitted alternately at different times, the voices are not mixed, and as a result, the user can easily listen to the voices. However, in the heterogeneous squealing method, since a voice is emitted from a speaker or the like installed at a specific place like a pedestrian crossing, the user cannot be guided outside the specific place.

Therefore, in this embodiment, the guidance guidance is not limited to the above specific position, and it is possible to intuitively grasp the correct traveling direction and how to return when leaving the route as in the case of different squealing methods. Describe the system.
<2. Schematic configuration of guidance system>
First, with reference to FIG. 1, the schematic structure of the guidance system 1 which concerns on this embodiment is demonstrated. FIG. 1 is an explanatory diagram illustrating an example of a schematic configuration of the guidance system 1 according to the present embodiment. Referring to FIG. 1, the guidance and guidance system 1 includes a sensor 10, a voice output device 20, and a voice processing device 100.

(Sensor 10)
The sensor 10 is a device that detects the position of the user 3 (hereinafter referred to as “user position”). The sensor 10 may include, for example, a GPS (Global Positioning System) receiver. Alternatively, the sensor 10 receives position information by radio waves, infrared rays, or the like from an active marker or passive marker provided in an environment in which the user 3 moves as disclosed in, for example, Japanese Patent Application Laid-Open No. 2003-91794. A machine may be provided. Here, the sensor 10 acquires the position information by acquiring the ID instead of the position information from the active marker or the passive marker by the receiver, and accessing the server device that stores the correspondence relationship between the ID and the position information. It may be possible.

  The sensor 10 further detects, for example, the orientation of the user 3 (hereinafter referred to as “user orientation”). The sensor 10 may include any of a gyro sensor, a geomagnetic sensor, and an acceleration sensor. Alternatively, the sensor 10 may include the receiver that receives information from the active marker or the passive marker, and may detect the direction of the user 3 from the positional relationship between the receiver and the active marker or the passive marker.

  Referring to FIG. 1, as an example, when the sensor 10 includes a GPS receiver and an acceleration sensor, the sensor 10 is mounted on the waist of the user 3. When the sensor 10 is a receiver that receives information from an active marker or a passive marker, the sensor 10 may be formed in the same shape as the white cane used by the user 3 and carried by the user 3.

(Audio output device 20)
The audio output device 20 is a device that outputs audio of audio data created by the audio processing device 100. The audio output device 20 is, for example, headphones that emit 2-channel stereo sound. The audio output device 20 acquires, for example, an analog audio signal obtained by digital / analog conversion (hereinafter referred to as “D / A conversion”) of audio data from the audio processing device 100, and based on the analog audio signal. To output sound. The audio output device 20 may acquire the audio data itself from the audio processing device 100 and perform D / A conversion on the audio data. Referring to FIG. 1, as an example, the audio output device 20 is worn on the head of the user 3 in the case of the headphones.

(Speech processor 100)
The voice processing device 100 is a device that creates voice data for enabling an intuitive grasp of the correct traveling direction and how to return when leaving a route. The audio processing device 100 is connected to the sensor 30 and the audio output device 20 by wire or wireless. For example, the voice processing apparatus 100 creates voice data based on the user position and the user orientation detected by the sensor 10. Then, the audio processing device 100 outputs, for example, an analog audio signal obtained by D / A converting audio data to the audio output device 20. When the audio output device 20 performs D / A conversion on the audio data, the audio processing device 100 may output the audio data to the audio output device 20. Note that the specific configuration of the voice processing apparatus 100 and the specific voice processing by the voice processing apparatus 100 will be described in <3. Configuration of speech processing apparatus> and <4. Processing flow> will be described.

  As mentioned above, although an example of the structure of the guidance system 1 which concerns on embodiment of this invention with reference to FIG. 1 was demonstrated, the structure of the guidance system 1 is not restricted to this. For example, the sensor 10, the sound output device 20, and the sound processing device 100 have been described as separate devices that are physically separated, but any two or more of these devices are physically integrated. It may be.

<3. Configuration of speech processing apparatus>
Next, an example of the configuration of the speech processing apparatus 100 according to the present embodiment will be described with reference to FIGS. FIG. 2 is a block diagram illustrating an example of the configuration of the speech processing apparatus 100 according to the present embodiment. Referring to FIG. 2, the voice processing apparatus 100 includes a detection unit 110, a target input unit 120, a storage unit 130, a guidance information creation unit 140, a sound source setting unit 150, a voice creation unit 160, and a voice output unit 170.

(Detector 110)
The detection unit 110 detects the user position _Pu . More specifically, the detection unit 110 is, for example, by obtaining the user position detected by the sensor 10 from the sensor 10 detects the user position P _u. The user position P _u is represented by, for example, plane coordinates (x _u , _yu ) consisting of an x coordinate and ay coordinate. For example, when the sensor 10 includes a GPS receiver, the x coordinate and the y coordinate may be latitude and longitude, respectively. Note that the user position P _u may be represented by spatial coordinates (x _u , _yu , z _u ) made up of an x coordinate, ay coordinate, and a z coordinate. In this case, the z coordinate may be a value indicating a height such as an altitude.

Moreover, the detection part 110 further detects a user's direction, for example. More specifically, the detection unit 110 detects the direction of the user by acquiring the direction of the user detected by the sensor 10 from the sensor 10, for example. The direction of the user here is, for example, the front direction of the user, and is represented by an angle θ _u formed by the front direction of the user and a predetermined direction. FIG. 3 is an explanatory diagram for explaining an example of a detected user orientation. Referring to FIG. 3, for example, when the user position P _u is represented by the plane coordinates (x _u , y _u ), the user orientation is the angle θ formed by the positive x-axis direction and the user orientation. It is represented by _u .

(Target input unit 120)
The target input unit 120 acquires a final target position (hereinafter referred to as “final target position”) of movement by the user 3. Final target position is represented by, for example, user position P _u similar format. The target input unit 120 acquires the final target position in accordance with, for example, an input operation by the user 3. More specifically, the target input unit 120 presents the final target position candidate to the user 3 by voice, and acquires the candidate selected by the user 3 using the operation unit such as a button, switch, or lever as the final target position. May be. Alternatively, the target input unit 120 may detect a number selected by the user 3 using the operation unit and acquire the number as the plane coordinates of the final target position. Alternatively, the target input unit 120 may recognize the collected voice of the user 3 and acquire a final target position corresponding to the recognized voice. The target input unit 120 may request the user 3 to perform an input operation in response to a start instruction from the user using the operation unit, or the user position _Pu may specify a subway ticket gate exit, a pedestrian crossing end, or the like. The user 3 may be requested to perform an input operation. On the other hand, instead of the input operation by the user 3, the target input unit 120 may automatically acquire the final target position. For example, when the user position _Pu is a specific position such as a subway ticket gate exit or a crosswalk end, the target input unit 120 automatically acquires a predetermined position corresponding to the specific position as the final target position. May be.

(Storage unit 130)
The storage unit 130 stores information that should be temporarily or permanently stored in the speech processing apparatus 100. The storage unit 130 stores, for example, map information in a space where the user moves, and a route created by the guidance information creation unit 140 described later. The storage unit 130 also stores sample audio data for creating audio data. The storage unit 130 may be a magnetic recording medium such as a hard disk, an EEPROM (Electrically Erasable and Programmable Read Only Memory), a flash memory, a MRAM (Magnetoresistive Random Access Memory), It may be a non-volatile memory such as FeRAM (Ferroelectric Random Access Memory) or PRAM (Phase change Random Access Memory).

(Guidance information creation unit 140)
Guide information creating unit 140 sets the target position P _g should guide the user 3, the start position P _s of guidance to the target position. For example, the guidance information creation unit 140 creates a route to the final target position acquired by the target input unit 120 based on the map information stored in the storage unit 130, and based on the route, the target position _Pg and setting the start position _{P s.} Hereinafter, this point will be described more specifically with reference to FIG.

FIG. 4 is an explanatory diagram for explaining an example of a route created by the guidance information creation unit 140. As shown in 4-1 of FIG. 4, as an example, the final target position 43 is a position that can be reached from the user position 41 at the time of route creation by linear movement. In this case, the guidance information creation unit 140 creates a route including the user position 41 and the final target position 43 when the route is created. The guide information creation unit 140, for example, sets the user position 41 during route generation starting at P _s, sets the final target position 43 as the target position P _g.

As shown in 4-2 of FIG. 4, as another example, the final target position 43 is a position that can be reached from the user position 41 at the time of route creation by changing the traveling direction. In this case, the guidance information creation unit 140 creates a route including the user position 41 at the time of route creation, the final target position 43, and a position 45 on the route whose traveling direction should be changed. The guidance information creation unit 140 first sets, for example, the user position 41 at the time of route creation as the start position P _s1 , and sets the position 45a on the route that can be reached by linear movement from the start position P _s1 as the target position P. Set as _g1 . Thereafter, when the user 3 reaches the target position P _g1 that is the position 45a on the route, the guidance information creation unit 140 sets the position 45a on the route as a new start position P _s2 , and starts from the start position P _s2. A position 45b on the path that can be reached by linear movement is set as a new target position _Pg2 . Thereafter, similarly, the guidance information creation unit 140 sets a position 45b on the route as a new start position _Ps3 , and sets a final target position 43 as a new target position _Pg3 . Note that the guidance information creation unit 140 stores the created route in the storage unit 130, for example.

The guide information creation unit 140, for example, determines whether the user 3 reaches the target position P _g. Guide information creating unit 140, when the case where the target position P _g and the user position P _u match, or the distance between the target position P _g and the user position P _u is equal to or less than a predetermined threshold value, the user 3 the target position It may be determined that _Pg has been reached. The guide information creation unit 140, also based on the user's orientation, it may be determined that the user 3 has reached the target position P _g. For example, the guide information creating unit 140 is a is the distance is equal to or less than a predetermined threshold value, and if the user of the orientation coincides with the direction of the next target position P _g, i.e. the user 3 toward the next target position P _g If you start the process proceeds Te, it may determine that the user 3 has reached the target position P _g.

(Sound source setting unit 150)
The sound source setting unit 150 includes a first virtual sound source position corresponding to the start position P _s (hereinafter referred to as “start side virtual sound source position”) and a second virtual sound source position corresponding to the target position P _g (hereinafter referred to as “start virtual sound source position”). (Referred to as “target-side virtual sound source position”). Sound source setting unit 150, for example, set the start position P _s that is set by the guide information creation section 140 as a starting virtual sound source position, target virtual sound source position set target position P _g by the guide information creation portion 140 Set as. Hereinafter, an example in which the start side virtual sound source position and the target side virtual sound source position are set in this way will be described. In order to easily understand that this setting is a prerequisite, the start side virtual sound source position is described as a start side virtual sound source position P _s , and the target side virtual sound source position is described as a target side virtual sound source position P _g .

(Voice creation unit 160)
The sound creation unit 160 uses the user position P _u detected by the detection unit 110 to generate first sound data (hereinafter referred to as “start side sound”) that the user 3 perceives the start side virtual sound source position P _s as the sound source position. And second audio data (hereinafter referred to as “target-side audio data”) in which the user 3 perceives the target-side virtual sound source position _Pg as a sound source. Also, each of the sound creation unit 160, for example, the each initiator audio data or the target-side audio data, and user position P _u or user orientation, starting virtual sound source position P _s or target virtual sound source position P _g Create based on the relative relationship. Hereinafter, this point will be described more specifically with reference to FIG.

FIG. 5 is an explanatory diagram for explaining creation of voice data by the voice creation unit 160. Referring to FIG. 5, for example, the sound creation unit 160 includes the distance d ₁ between the user position P _u and the start-side virtual sound source position P _s , the user orientation (the user's front direction), and the start-side virtual sound source position P _s. An angle θ ₁ formed by the directions is calculated. Then, the voice creation unit 160 creates voice data in which the position separated by the distance d _{1 in} the direction shifted by the angle θ ₁ from the front direction is perceived as the sound source as the start side voice data. Similarly, the voice creation unit 160, for example, the distance d ₂ between the user position P _u and the target-side virtual sound source position P _g and the direction of the user (the front direction of the user) and the direction of the target-side virtual sound source position P _g The formed angle θ ₂ is calculated. Then, the sound creation unit 160 creates sound data in which the position separated by the distance d _{2 in} the direction shifted by the angle θ ₂ from the front direction is perceived as the sound source as the target side sound data. That is, the relative relationship, for example, the distance between the user position P _u start and virtual sound source position P _s d _1, user position P _u and the distance d ₂ between the target virtual sound source position P _g, and the user's orientation It includes an angle θ ₁ formed with the direction of the start-side virtual sound source position P _s , or an angle θ ₂ formed between the user direction and the direction of the target-side virtual sound source position P _g .

A method for calculating the distances d ₁ and d ₂ and the angles θ ₁ and θ ₂ will be described. Referring to FIG. 5, first, the sound creation unit 160 first detects the user position P _u and the user orientation θ _u detected by the detection unit 110, and the start side virtual sound source position P _s set by the sound source setting unit 150. The target side virtual sound source position _Pg is acquired. Here, the user position P _u , the start-side virtual sound source position P _s, and the target-side virtual sound source position P _g are respectively represented by plane coordinates (x _u , _yu ), (x _s , y _s ), and (x _g , y _g). ). Then, the voice creation unit 160 calculates the distances d ₁ and d ₂ by the following equations (1) and (2).

（１）

（２）

また、音声作成部１６０は、以下の式（３）および（４）により、ユーザ位置Ｐ_ｕから開始側仮想音源位置Ｐ_ｓへの方向θ_ｓ、およびユーザ位置Ｐ_ｕから目標側仮想音源位置Ｐ_ｇへの方向θ_ｇを算出する。

(1)

(2)

The audio creation unit 160, following the equation (3) and (4), the direction theta _s from the user position _{P u} to start virtual sound source position _{P s,} and user location _{P u} target virtual sound source position from the P to calculate the direction θ _g to _g.

（３）

（４）

そして、音声作成部１６０は、以下の式（５）および（６）により、角度θ_１およびθ_２を算出する。

(3)

(4)

Then, the voice creation unit 160 calculates the angles θ ₁ and θ ₂ by the following equations (5) and (6).

（５）

（５）

(5)

(5)

In addition, the voice creation unit 160 uses, for example, the following sound source localization method, start side voice data in which a position separated by a distance d _{1 in} a direction shifted by an angle θ ₁ from the front direction is perceived as a sound source, and the front direction And target side audio data in which a position separated by a distance d _{2 in} a direction deviated by an angle θ ₂ from the target is perceived as a sound source. The voice creation unit 160 creates the start side voice data and the target side voice data in a stereo format.

As a first example, the voice creation unit 160 creates start-side voice data and target-side voice data by convolution of a head related transfer function (HRTF). HRTF is a function representing the transfer characteristic of sound from the sound source to the ear. For example, the HRTF can be obtained by measuring the impulse response of sound emitted from a discretely arranged sound source with a dummy head and a microphone attached to the ear. The sound creation unit 160 can create start-side sound data by convolving the first sample sound data stored in the storage unit 130 with the HRTF corresponding to the angle θ ₁ and the distance d ₁ in the time domain. it can. Similarly, the voice creation unit 160 creates target side voice data by convolving the HRTF corresponding to the angle θ ₂ and the distance d ₂ with the second sample voice data stored in the storage unit 130 in the time domain. can do.

As a second example, the sound creation unit 160 creates start-side sound data and target-side sound data by adjusting the balance of volume between channels in the stereo format. For example, when each audio data is generated in the left and right two-channel stereo format, the audio generation unit 160 adjusts the volume balance between the left channel and the right channel according to the angle θ ₁ and the distance d _1. Thus, the start side audio data is created. Similarly, the voice creation unit 160 creates target-side voice data by adjusting the volume balance between the left channel and the right channel according to the angle θ ₂ and the distance d ₂ .

Heretofore, the creation method of the start side voice data and the target side voice data by the voice creation unit 160 has been described. Next, with reference to FIGS. 6 to 8, it will be described how the audio data to the user 3 moves from the start position P _s in the target position P _g is created.

Figure 6 is an explanatory view for the user 3 illustrating the creation of audio data up to point toward the target position P _g. Referring to FIG. 6, 6-1, induction of the user 3 to the target position P _g is started from the start position P _s. Then, over a period of from 6-1 6-3, user 3, changing the direction so that the direction of the user of the orientation and the target position P _g coincide. Along with this, the angle θ ₂ formed by the user's direction and the direction of the target position P _g (that is, the target sound source position P _g ) changes, so that the sound creation unit 160 is shifted from the front direction by the angle θ ₂ . position apart direction by a distance d ₂ to create a target-side audio data that is perceived as a sound source at any time. Note that since the start position P _s (that is, the start-side sound source position P _s ) and the user position P _u coincide with each other, there is no angle θ ₁ formed by the user direction and the start-side virtual sound source position P _s . However, as shown in FIG. 6, the sound creation unit 160 is, for example, regarded as the direction of the start-side virtual sound source position P _s The user of the orientation of the induction starting 6-1. That is, the voice creation unit 160 determines that any position in a direction shifted by an angle θ ₁ formed by the current user direction and the user direction at the start of guidance (the direction of the starting virtual sound source position P _s ) is a sound source. Create the start side audio data perceived as Thus, the user of the orientation of at induction start is regarded as the direction of the start-side virtual sound source position P _s at the start position P _s, the user 3 at any time in the direction of the induction at the beginning of the user at the start position P _s It becomes possible to turn around.

Figure 7 is an explanatory diagram for explaining the creation of sound data when the user 3 linearly moves toward the target position P _g. Referring to FIG. 7, at 7-1, the user 3 starts moving to the target position P _g. Then, over a period of from 7-1 7-3, the user 3 is moved toward the target position _{P g.} Accordingly, since the distance d ₂ between the user position P _u and the target position P _g (that is, the target-side sound source position P _g ) changes, the sound creation unit 160 has a position separated by the distance d _{2 in the} front direction. Create target-side audio data perceived as a sound source. In addition, since the distance d ₁ between the user position P _u and the start position P _s (that is, the start-side sound source position P _s ) changes, the sound creation unit 160 uses a position separated by the distance d _{1 in the} back direction as a sound source. Create perceived starter audio data as needed. As described above, when creating the voice data as needed, the voice creation unit 160 changes the volume of the start side voice data and the target side voice data according to the distance d ₁ and the distance d ₂ , for example. As described above, the user 3 can reach the target position P _g from the start position P _s be moved as shown in FIGS. However, as described below, the user 3, when attempts to return to the start position P _s, there is a possibility to change the orientation in the middle of the movement, such as when out of the path.

FIG. 8 is an explanatory diagram for explaining creation of audio data when the user 3 changes direction during movement. Referring to FIG. 8, 8-1, user 3 stops moving to the target position P _g. And from 8-1 to 8-3, the user 3 changes direction to the right side. Along with this, the angle θ ₂ formed by the user's direction and the direction of the target position P _g (that is, the target sound source position P _g ) changes, so that the sound creation unit 160 is shifted from the front direction by the angle θ ₂ . position apart direction by a distance d ₂ to create a target-side audio data that is perceived as a sound source at any time. Similarly, since the angle θ ₁ formed by the direction of the user and the direction of the start position P _s (that is, the start-side sound source position P _s ) changes, the sound creation unit 160 is shifted from the front direction by the angle θ _1. Starting side audio data in which a position separated by a distance d ₁ is perceived as a sound source is generated as needed.

The creation of such initiator audio data and the target-side audio data, even if it is not installed device that emits a sound such as a speaker in a specific position, the start position P _s and the target position P _g is present per any It is possible to make the user 3 intuitively grasp this. In addition, since the voice of these voice data that the user 3 hears changes according to the movement of the user, the user can check whether the voice is changing in the correct traveling direction or not away from the route. Intuitive grasp at any time. Moreover, since the initiating audio data not target side audio data only are also created, the user can return to the correct path and the start position P _s. That is, even in a place where a device that emits voice is not installed, it is possible to intuitively grasp the correct direction of travel and how to return when the route is away from the route, as in the case of different squealing methods.

The start side audio data and the target side audio data created as described above have different audio patterns, for example. The sound pattern here is, for example, the type of sound. The start side audio data has, for example, a type of audio “Cuckoo”, and the target side audio data has, for example, a type of audio “Cuckoo”. In order to provide more specific information, the start side audio data has a sound type of “start position” or “back”, and the target side audio data has an audio type of “target position” or “front”. You may have a kind. The sound pattern is not limited to the type of sound, and may be, for example, the pitch of the sound, the tempo, or the like. Thus, if the voice patterns of the start side voice data and the target side voice data are different from each other, the user 3 can distinguish between the voices of the two. As a result, the user 3, which is the start position P _s, which is possible to identify whether the target position P _g.

  As described above, the voice creation unit 160 creates the start side voice data and the target side voice data, respectively. Then, the voice creation unit 160 arranges the created start-side voice data and target-side voice data at different positions on the time axis. FIG. 9 is an explanatory diagram for explaining the arrangement of audio data on the time axis by the audio creation unit 160. Referring to FIG. 9, as shown in the time zone t1, the voice creation unit 160 alternately arranges the start side voice data A and the target side voice data B on the time axis, for example. Alternatively, as shown in the time zone t2, the voice creation unit 160 may arrange the start side voice data A and the target side voice data B at different frequencies, for example. Thus, if the start side audio data and the target side audio data are arranged at different positions on the time axis, the start side audio data and the target side audio data are not mixed. As a result, the user 3 can easily distinguish the voice of the start side voice data and the voice of the target side voice data. Note that the voice creation unit 160 arranges the start-side voice data and the target-side voice data at different positions on the time axis by setting a parameter indicating the voice output frequency or output timing, for example.

(Audio output unit 170)
The audio output unit 170 generates an analog audio signal by performing D / A conversion on the start side audio data and the target side audio data generated by the audio generation unit 160, and outputs the analog audio signal to the audio output device 20. . The audio output unit 170 outputs an analog audio signal according to the arrangement of audio data on the time axis by the audio generation unit 160. When the audio output device 20 performs D / A conversion on the audio data, the audio output unit 170 may output the audio data itself to the audio output device 20.

  As mentioned above, with reference to FIGS. 2-10, although the example of the structure of the speech processing apparatus 100 which concerns on this embodiment was demonstrated, the structure of the speech processing apparatus 100 which concerns on this embodiment is not limited to this example.

(Modification)
- creation of audio data, for example, voice creation unit 160, each of the start-side audio data and the target-side audio data, and user position P _u or user orientation, starting virtual sound source position P _s or target virtual sound source position P You may produce by the sound volume or audio | voice pattern according to the relative relationship with each of _g . As already explained, the relative relationship, for example, the user position P the distance d ₁ between _u and start virtual sound source position P _s, the user position P _u and the distance d ₂ between the target virtual sound source position P _g, It includes an angle θ ₁ formed by the user direction and the direction of the start-side virtual sound source position P _s , or an angle θ ₂ formed by the user direction and the direction of the target-side virtual sound source position P _g . The sound pattern is, for example, a sound type, a sound pitch, a tempo, and the like.

As a specific example, referring to FIG. 7 again, for example, the voice creation unit 160 increases the voice tempo in the start-side voice data or the target-side voice data as the distance d ₁ or the distance d ₂ decreases. Also good. Alternatively, the voice creation unit 160 may change the voice tempo in the start-side voice data and the voice tempo in the target-side voice data according to the ratio between the distance d ₁ and the distance d ₂ . Further, referring to FIG. 6 or FIG. 8 again, the sound creation unit 160 may increase the volume in the start side sound data or the target side sound data as the angle θ ₁ or the angle θ ₂ decreases, for example. Alternatively, the voice creation unit 160 may change the volume in the start side voice data and the volume in the target side voice data in accordance with the ratio between the angle θ ₁ and the angle θ ₂ . Alternatively, the voice creation unit 160 determines that the angle θ ₁ or the angle θ ₂ is 0 ° or a value smaller than a predetermined threshold, that is, the user 3 is facing the start position P _s or the target position P _g . In addition, the voice type in the start side voice data or the target side voice data may be a voice type different from other cases.

As another specific example, the voice creation unit 160 may create each of the start-side voice data and the target-side voice data with guidance information (that is, the type of voice) according to the relative relationship. For example, the guidance information creation unit 140 creates guidance information regarding the start position P _s or the target position P _g and sample voice data in which the guidance information is voiced. Then, the voice creation unit 160 creates start-side voice data or target-side voice data using the sample voice data created by the guidance information creation unit 140. The guidance information is, for example, information "d ₂ meters target, the angle theta ₂ degrees". Or, the guidance information may be information indicating what is in the start position P _s or the target position P _g.

As described above, by creating a starting side audio data and the target-side audio data in sound or voice patterns corresponding to the relative relationship, to recognize the position of the start position P _s or the target position P _g more easily To the user 3. For example, user 3, that the direction of the start position P _s or the target position P _g can be easily recognized, or easily recognize whether the approaching start position P _s or the target position P _g it can.

-Arrangement of audio data on time axis In addition, the audio creation unit 160 arranges each of the start side audio data and the target side audio data at different positions on the time axis at a frequency according to the relative relationship. Also good.

Specifically, the sound creation unit 160, for example, as the distance d ₁ or the distance d ₂ is smaller, may be arranged initiator audio data or the target side voice data more frequently. Alternatively, the voice creation unit 160 may arrange the start side voice data or the target side voice data at a higher frequency, for example, as the angle θ ₁ or the angle θ ₂ decreases. Alternatively, the voice creation unit 160 may change the frequency of the start-side voice data and the target-side voice data according to the ratio between the distance d ₁ and the distance d ₂ or the ratio between the angle θ ₁ and the angle θ _2. Good. Referring again to FIG. 9, for example, when the ratio d ₁ / d ₂ increases from the time zone t ₁ to the time zone t ₂ as a result of the user 3 approaching the target position P _g from the start position P _s , The voice creation unit 160 may increase the frequency of the target side voice data B.

As described above, by changing the frequency to be arranged on the time axis corresponding to the relative relationship allows the user 3 to recognize the position of the start position P _s or the target position P _g more easily .

- Setting of the virtual sound source position also sound source setting unit 150, the setting may be setting the position other than the start position P _s as a starting virtual sound source position and the position other than the target position P _g as the target virtual sound source positions May be. Hereinafter, a specific modification of the setting of the virtual sound source position will be described with reference to FIG.

FIG. 10 is an explanatory diagram for describing a modified example of the setting of the virtual sound source position. As an example, as illustrated in 10-1 and 10-2 of FIG. 10, the sound source setting unit 150 uses any position 51 between the start position P _s and the user position _Pu as a start-side virtual sound source position. It may be set. Similarly, the sound source setting portion 150 may set one of the positions 53 between the target position P _g and the user position P _u as the target virtual sound source positions.

As another example, as shown in 10-3 and 10-4 in FIG. 10, the sound source setting unit 150, a position on the path from the start position _{P s} to the target position _{P g,} the user position _{P u} may set the position 55 it should pass to return to the start position P _s as a starting virtual sound source position from. Similarly, the sound source setting unit 150 is a position on the path from the start position P _s to the target position P _g that should pass through to reach the target position P _g from the user position _Pu. You may set as a virtual sound source position.

As yet another example, the sound source setting unit 150 uses any position in the vicinity of the start position P _s and any position in the vicinity of the target position P _{g as} the start side virtual sound source position and the target side virtual sound source position, respectively. It may be set.

  As another viewpoint, the sound source setting unit 150 may set the start-side virtual sound source position and the target-side virtual sound source position in a three-dimensional space instead of on a two-dimensional plane. In this case, the voice creation unit 160 may create start-side voice data or target-side voice data in which a position higher or lower than the user 3 is perceived as a sound source using HRTF. According to such audio data, the user 3 can intuitively grasp the target position, for example, downstairs or upstairs.

<4. Process flow>
Hereinafter, an example of the position estimation process according to the present embodiment will be described with reference to FIG. FIG. 11 is a flowchart illustrating an example of a schematic flow of the position estimation process according to the present embodiment.

First, in step S201, the target input unit 120 acquires a final target position. In step S203, the detection unit 110 detects the user position _Pu and the user orientation. In step S205, the guidance information creation unit 140 creates a route from the user position _Pu to the final target position based on the map information stored in the storage unit 130.

Next, in step S207, the guide information creating unit 140, based on the route that was created, sets the target position P _g should guide the user 3, the start position P _s of guidance to the target position. In step S209, the sound source setting unit 150, the initiator virtual sound source position corresponding to the start position P _s, and sets the target virtual sound source position corresponding to the target position P _g.

Next, in step S211, the voice creation unit 160 uses the user position P _u detected by the detection unit 110 to start side voice data in which the user 3 perceives the start side virtual sound source position P _s as a sound source position, Then, target side audio data in which the target side virtual sound source position _Pg is perceived by the user 3 as a sound source is created. In step S213, the voice creation unit 160 arranges the start side voice data and the target side voice data at different positions on the time axis. In step S215, the audio output unit 170 generates an analog audio signal by performing D / A conversion on the start side audio data and the target side audio data generated by the audio generation unit 160, and converts the analog audio signal into audio. Output to the output device 20.

Next, in step S217, the detection unit 110 detects the user position _Pu and the user orientation. Next, in step S219, the voice creation unit 160 determines whether the user position _Pu or the user orientation has changed. If the user position _Pu or the user orientation has changed, the process proceeds to step S221. Otherwise, the process returns to step S213.

In step S221, the guide information creating unit 140 determines whether the user 3 reaches the target position _{P g.} If the user 3 long reached the target position _{P g,} the process proceeds to step S223. Otherwise, the process returns to step S211.

In step S221, the guidance information creation unit 140 determines whether or not the target position _Pg is the final target position. If the target position _Pg is the final target position, the process ends. Otherwise, the process returns to step S207.

  As described above, according to the embodiment of the present invention, according to the present embodiment, even if a device that emits a sound is not installed at a specific position, it is possible to start from a correct traveling direction and route as in the case of different types of squealing. It is possible to intuitively understand how to return when leaving.

  In addition, although preferred embodiment of this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to the example which concerns. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  Further, each step in the voice processing of this specification does not necessarily have to be processed in time series in the order described in the flowchart. For example, each step in the audio processing may be processed in an order different from the order described as the flowchart, or may be processed in parallel.

  Further, it is possible to create a computer program for causing hardware such as a CPU, a ROM, and a RAM built in the voice processing apparatus 100 to perform the same functions as the components of the voice processing apparatus 100. A storage medium storing the computer program is also provided.

DESCRIPTION OF SYMBOLS 1 Guide guidance system 3 User 10 Sensor 20 Audio | voice output apparatus 100 Audio | voice processing apparatus 110 Detection part 120 Target input part 130 Memory | storage part 140 Guidance information creation part 150 Sound source setting part 160 Audio | voice creation part 170 Audio | voice output part

Claims (10)

A detection unit for detecting a user position;
A first virtual sound source position corresponding to a start position which is the user position detected by the detection unit at the time when guidance to the user is started, and a target position of the guidance or between the user position and the target position a sound source setting unit that sets a second virtual sound source position corresponding to the position existing on the path,
Using the user position detected by the detection unit, first audio data in which the first virtual sound source position is perceived by the user as a sound source position, and the second virtual sound source position is perceived by the user as a sound source. A voice creation unit for creating second voice data to be played,
Equipped with a,
The first audio data and the second audio data have different audio patterns,
The sound creation unit place the first audio data and the second audio data at different positions on the time axis, the sound processing apparatus. The detection unit further detects the orientation of the user;
The voice creation unit converts each of the first voice data or the second voice data into the user position or the user orientation, the first virtual sound source position, or the second virtual sound source position. created based on the relative relationship between the sound processing device according to claim 1. The voice processing device according to claim 2 , wherein the voice creation unit creates each of the first voice data and the second voice data with a volume or a voice pattern corresponding to the relative relationship. The sound creation unit, each of the first audio data and the second audio data is placed in different positions on the time axis at a frequency corresponding to the relative relationship, claim 2 or 3 The speech processing apparatus according to one item. The relative relationship is the distance between the user position and the first virtual sound source position or the second virtual sound source position, or the direction of the user and the direction of the first virtual sound source position or the first virtual sound source position. including an angle between the direction of the second virtual sound source position, the sound processing apparatus as claimed in any one in claim 2-4. The sound creation unit, the first audio data and the second sound data to create a stereo format, audio processing apparatus according to any one of claims 1-5. The speech processing apparatus according to claim 6, wherein the speech creation unit creates the first speech data and the second speech data by convolution of a head related transfer function. A step of detecting a user position by the detection unit ;
The first virtual sound source position corresponding to the start position, which is the user position detected by the detection unit when the sound source setting unit starts guidance to the user , and the target position of the guide or the user position and the target and setting a second virtual sound source position corresponding to the position existing on the path between the position,
The sound source preparation section is detected by the detection unit with a user position, the first audio data have been the first virtual sound source position set by the setting in the sound source unit is perceived by the user as a sound source position, and the Creating second voice data having a voice pattern different from the first voice data, and the second virtual sound source position set by the sound source setting unit being perceived by a user as a sound source;
The sound source creation unit placing the first audio data and the second audio data at different positions on a time axis;
Including a voice processing method. Computer
A detection unit for detecting a user position;
A first virtual sound source position corresponding to a start position which is the user position detected by the detection unit at the time when guidance to the user is started, and a target position of the guidance or between the user position and the target position a sound source setting unit that sets a second virtual sound source position corresponding to the position existing on the path,
Using the user position detected by the detection unit, the first virtual sound source position is perceived by the user as a sound source position, and has a sound pattern different from the first sound data. sound creation aforementioned second virtual sound source position is to create a second audio data that is perceived to the user as a sound source, arranging the first audio data and the second audio data at different positions on the time axis And
Program to function as. A guidance guidance system including a sensor, a voice output device and a voice processing device,
The voice processing device
A detection unit for detecting a user position based on an input from the sensor;
A first virtual sound source position corresponding to a start position which is the user position detected by the detection unit at the time when guidance to the user is started, and a target position of the guidance or between the user position and the target position a sound source setting unit that sets a second virtual sound source position corresponding to the position existing on the path,
Using the user position detected by the detection unit, first audio data in which the first virtual sound source position is perceived by the user as a sound source position, and the second virtual sound source position is perceived by the user as a sound source. A voice creation unit for creating second voice data to be played,
With
The first audio data and the second audio data have different audio patterns,
The voice creation unit arranges the first voice data and the second voice data at different positions on the time axis,
The sound output device outputs the sound of the first sound data and the sound of the second sound data;
Guide guidance system.

Images