WO2002082413A1

WO2002082413A1 - Voice guide system based on cry and song

Info

Publication number: WO2002082413A1
Application number: PCT/JP2002/002676
Authority: WO
Inventors: Kazuhiro Nakadai; Ken-Ichi Hidai; Hiroshi Okuno; Hiroaki Kitano
Original assignee: Japan Science And Technology Corporation
Priority date: 2001-04-05
Filing date: 2002-03-20
Publication date: 2002-10-17
Also published as: JP2002304191A; JP3590869B2

Abstract

A voice guide system for providing an explanation of an animal or a bird to a user based on the cry of an animal or the song of a bird, comprising a microphone array (11) of a plurality of fixed microphones (11a, 11b, ...), at least one headset (20) comprising a GPS (22), a movable microphone (21) and a headphone (23), a section (12) for locating and separating the sound source of each cry or song from the sound signal of each fixed microphone and movable microphone, a decision section (13) for specifying the kind of animal or bird uttering the cry or song from the separated sound data of each sound source, and a guide generating section (14) for converting the absolute coordinates position of a sound source located at the sound source separating/locating section into relative coordinates position based on the GPS positioning information of each headset and generating explanation data concerning the headset from the kind of animal or bird specified at the decision section and then delivering the explanation data to the headset.

Description

Description Voice-guided voice guidance system

The present invention relates to a sound guide system based on a call that provides a user with commentary on the animal or bird based on the sound of the animal or bird in a facility such as a natural park. Technology background

In general, in natural parks and the like, users can observe the natural appearance of animals and birds while appropriately moving within the premises of the natural park and the like. To provide users with information on animals and birds, for example, images such as pictures and photographs of animals and birds, information boards with explanations on them, and pamphlets are prepared.

However, it is usually difficult for users to discriminate the sounds of animals and birds. For this reason, the user cannot discriminate the kind of the animal or bird even if the call is heard. Therefore, especially in the case of wild birds, even if an attempt is made to find the shape of the wild bird, if the wild bird is unknown, it cannot be easily found because the wild bird itself is small.

In view of the above points, an object of the present invention is to provide a voice guidance system based on sounds of animals and birds, which provides a user with an explanation of the animals and birds based on the sounds of the sounds. Disclosure of the invention

According to the present invention, there is provided a microphone array including a plurality of fixed microphones appropriately arranged on a site of a facility such as a natural park, a GPS, a movable microphone, and a microphone carried by a user of the facility. At least one headset consisting of headphones, each fixed microphone of the microphone array and a movable microphone of each headset Sound source separation for sound source localization and separation of sound data for each sound source from the sound signal from the sound source.From the sound source localization part and the sound data for each sound source separated by the above sound source separation and sound source localization part. A discriminator that identifies the type of animal that made the call, etc., and the absolute coordinate position of the sound source localized by the sound source separation '

A guide to convert to the relative coordinate position based on the position information by S, create commentary data on the headset based on the kind of animal etc. specified by the discriminator, and output it to the headphone of the headset The present invention is achieved by a voice-based voice guidance system, which comprises:

Preferably, the sound guide system based on a call according to the present invention preferably includes a receiving unit in which the sound source separation and sound source setting unit receives position information by GPS from each headset and an acoustic signal by a movable microphone. The guide creation unit includes a transmission unit for transmitting the commentary data, and each of the headsets transmits the position information by GFS and the acoustic signal from the movable microphone to the reception unit, and transmits the solution from the transmission unit. And a transmission / reception unit for receiving the description data.

Preferably, the sound guide system based on a call according to the present invention is arranged such that the sound source separation and sound source localization unit performs a mathematical solution using a general microphone array based on acoustic signals from each fixed microphone and each movable microphone. Performs sound source separation.

In the voice guidance system according to the cry of the present invention, preferably, the sound source separation 'sound source localization unit performs sound source separation by using a direction path finola.

In the voice guidance system based on a call according to the present invention, preferably, the sound source separation-sound source localization unit further performs sound source separation by German 3t component analysis.

In the voice guidance system according to the present invention, preferably, the discriminating unit refers to a type database including the sounds of the animals and the like, and specifies the type of the animals and the like from the sounds of the animals and the like.

In the voice guidance system based on a call according to the present invention, the discriminating unit preferably performs detailed discrimination such as sex discrimination of animals and the like and situation discrimination based on the type of the animal and the like and further by referring to various detailed databases. Perform

According to the above configuration, the fixed microphones of the microphone array and the movable microphones of each headset collect sounds of animals and birds in the premises such as a natural park, and the collected sounds are collected. From the sound signal, the sound source separation 'sound source localization unit separates the sound into the absolute coordinate position of the sound source of each call and sound data for each sound source. Then, the discriminating unit specifies the type of animal or the like that squeaked for each sound source from the sound data of each sound source た separated by the sound source separation 'sound source localization unit, and the guide creating unit determines The commentary data is created corresponding to the position information of each headset by the customer, and the output unit outputs the commentary data to the headphones of the headset.

This allows the user to obtain a commentary on the animal or bird that made the cry by listening to the commentary flowing from the headphone of the headset carried by the user. The animal, bird, etc. can be easily found based on the coordinate position.

The sound source separation and sound source localization unit has a receiving unit that receives the position information by GPS from each headset and the acoustic signal from the movable microphone, and the guide creation unit has a transmission unit that sends the commentary data. If each of the headsets includes a transmitting / receiving unit that transmits position information by GPS and an acoustic signal by a movable microphone to the receiving unit and receives commentary data from the transmitting unit, Each headset transmits the position information by GPS and the acoustic signal by the movable microphone to the sound source separation and sound source localization unit, and receives the commentary data from the guide creation unit. Can be transmitted and received. Therefore, the user can freely move around the premises such as a natural park while carrying the headset.

Sound source separationIf the sound source localization unit performs sound source separation based on acoustic signals from each fixed microphone and each movable microphone by a mathematical solution using a general micro phone array, it includes the movable microphone Sound source separation can be performed by such a mathematical solution using a microphone array in a broad sense.

When the sound source localization unit performs sound source separation by using a direction pass filter, sound source separation can be easily performed by using a direction pass filter.

When the sound source separation 'sound source localization unit further performs sound source separation by independent component analysis, sound source separation can be reliably performed by using independent component analysis. When the discriminating unit specifies the type of the animal or the like from the call of the animal with reference to the page database provided with the sound of the animal or the like, the call of the animal or the like to be discriminated is prepared in advance in a database. Therefore, the type of the animal or the like can be quickly identified from the call of the animal or the like.

When the discriminating unit performs detailed discrimination such as gender discrimination of animals and the like and situation discrimination based on the data of the animals and the like and further by referring to various detailed databases, a more detailed discrimination based on the call of the animals etc. Discrimination, that is, the gender of the animal, the situation under which the call is emitted, and the like can be performed. BRIEF DESCRIPTION OF THE FIGURES

The invention will be better understood on the basis of the following detailed description and the accompanying drawings, which show some embodiments of the invention. The embodiments shown in the accompanying drawings are not intended to specify or limit the present invention, but are described merely for facilitating the explanation and understanding of the present invention.

In the figure,

FIG. 1 is a block diagram showing an electric configuration of an embodiment of a voice guidance system based on bird calls according to the present invention.

FIG. 2 is a schematic perspective view showing the entire configuration of the voice guidance system of FIG. FIG. 3 is a block diagram showing a configuration of a head set in the voice guidance system of FIG.

FIG. 4 is a flowchart showing the operation of the voice guidance system of FIG. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows an electrical configuration of an embodiment of a voice guidance system based on a bird singing to which the present invention is applied. In FIG. 1, the voice guidance system 10 sounds from a plurality of fixed microphones 11 a, lib,... Appropriately arranged in the site 10 a of a facility such as a natural park shown in FIG. 2. Microphone array 11, sound source separation and sound source localization unit 12, discrimination unit 13, guide creation unit 14, and carried by facility users And head set 20.

As shown in Fig. 2, the microphone array 11 is composed of a plurality of fixed microphones 11a, lib, Have been. Each fixed microphone 11a, lib, · · · basically uses a directional microphone because it only has to collect the bird's singing from above. The sound signals of the fixed microphones '11a, lib, ··· are input to a sound source separation' sound source localization unit 12 described later. Note that each of the fixed microphones 11a, 11b, ... may be connected to the sound source separation and sound source localization unit 12 via a cable, or may be connected wirelessly.

The sound source separation and sound source localization section 12 is provided, for example, in a management building 1 Ob in the site 10 a of a facility such as a natural park, and includes a fixed microphone 1 of the microphone array 11. 1a, 1 lb, '--' and the sound of the bird based on the sound signals from the movable microphones of each headset 20 are detected, and the localization (position information to absolute coordinate position) of the sound source, The song of the bird is separated. At this time, the positions of the fixed microphones 11a, lib, and so on of the microphone array 11 were previously input to the sound source separation and sound source localization unit 12 and the position of each headset 20. Is detected by the position information by the GFS provided in the headset 20.

The sound source separation / sound source localization unit 12 includes a sound source separation unit 12a, a direction path filter 11b, and an ICA (Independent Component Analysis) unit 12c. Contains.

The sound source separation unit 12a uses a general microphone array based on sound signals from the fixed microphones 11a, lib,... 'Of the microphone array 11 and the movable microphones of each headset 20. It is configured to perform sound source separation by a mathematical solution.

The above-mentioned direction pass filter 12b is used to determine the phase difference between the binaural ears based on the sound signals from the fixed microphones 11a, lib, ... of the microphone array 11 and the movable microphones of each headset 20. Sound source separation is performed using IPD and binaural ¾ ^ difference IID. The ICA unit 12c performs independent component analysis based on the acoustic signals from the individual fixed microphones 11a, lib, In addition, based on the sound signal level of each microphone according to the unknown probability distribution, the sound signal of each sound source is restored by matrix calculation to separate the sound sources, and the absolute coordinate position of the sound sources is detected.

The discriminating unit 13 is also provided, for example, in the management building 10b in the facility site 10a of a natural park or the like, and is provided with the sound of birds separated by the sound source separation / sound source localization unit 1. The position information (absolute coordinate position) and the input of the position information of each headset 20 are used to specify the singing of each bird, and include a type discriminator 13a and a detailed discriminator 13b. In.

The type discriminating unit 13a refers to the type database 13c having various types of bird calls and identifies the bird from the bird calls, and provides information on the types and positions of the birds.

(Absolute coordinate position) Then, the position information of each head set 20 is sent to the guide creating unit 14.

Further, the detailed discriminator 13b, based on the bird's singing from the sound source separation and sound source localization unit 12 and the bird type information from the type discriminator 13a, determines the sound of the bird according to its gender, status, etc. Referring to the detailed database 13d with singing voices, the singing of the birds is used to make detailed judgments such as gender discrimination and situation discrimination, and the detailed information on the birds is sent to the guide creation unit 14.

The guide creation unit 14 is also provided in the management premises 10b, for example, in the site 10a of the facility such as a natural park, and the position of the discrimination unit 13 is determined for each headset 20. The position information (absolute coordinate position) is converted into the relative coordinate position of the bird position information based on the GFS position information of each headset 20 and the bird type information determined by the determination unit 13 Based on the detailed information and the detailed information, the commentary data on the headset 20 is created and transmitted to the headset 20 by the transmission / reception unit 4 shown in FIG.

Here, when creating the commentary data, the guide creating unit 14 uses the periphery of the head set 20, that is, the head set 20 based on the position information of each head set 20. The explanation is limited to the birdsong within the range where the user —Create evenings. In the illustrated case, the guide creating unit 14 is provided for each head set 20. However, the present invention is not limited to this, and one guide creating unit 14 corresponds to all the head sets 20. Commentary data may be created, and a guide creation unit 14 smaller in number than the number of headsets 20 is prepared. May be created. At least one headset 20 is provided so that users of facilities such as a natural park can carry it. As shown in Fig. 3, a movable microphone 21 and a GPS 2 2, a headphone 23, a transmission / reception unit 24, and a noise canceling circuit 25.

The movable microphone 21 detects a sound around the headset 20 and generates an acoustic signal. The above-mentioned GPS 22 has a known configuration, and detects the position of the head set 20 by receiving a radio wave from the GPS satellite to generate position information. The headphone 23 provides a commentary voice to the user based on the commentary data from the guide creation unit 14. The transmission / reception unit 24 transmits the sound signal from the movable microphone 21 and the position information from the GPS 22 to the sound source separation / sound source localization unit 12, It is designed to receive commentary data from 14. Further, the noise canceling circuit 25 controls the noise canceling so that the user using the headset 20 does not disturb the explanation sound from the headphones 23. .

The voice guide system 10 according to the embodiment of the present invention is configured as described above, and as shown in the flowchart of FIG. 4, a headset 2 is set for each user based on the birdsong in a facility such as a natural park. 0 provides commentary.

First, in FIG. 4, at step ST1, the fixed microphones lla, lib,... Of the microphone array 11 and the movable microphones 21 of the respective headsets 20 are used to make bird calls in facilities such as a natural park. And sends an acoustic signal to the sound source separation / sound source localization unit 12. Here, the positions of the fixed microphones 11a and 11b have been previously input to the sound source separation / sound source localization unit 12 and the position of the movable microphone 21 of each headset 20 is Headset 20 GPS location information is sound source Separation · This is known by being input to the sound source localization unit 12.

Next, in step ST2, the sound source separation / sound source localization unit 12 sends the sound from the fixed microphones lla, lib, ... of the microphone array 11 and the movable microphone 21 of each headset 20. Based on the signal, the sound source separation unit 12a performs sound source separation by a mathematical solution using a general microphone array 11 and a sound source using IFD and IID by a decimation pass filter 12b. By performing sound source separation by independent component analysis in the ICA unit 12c, the position information (absolute coordinate position) of the bird that emitted the squealing sound is detected, and Separate bird calls. Then, the sound source separation / sound source localization unit 12 outputs the bird's position information G »f coordinate position) and the bird's singing together with the position information of each headset 20 to the discrimination unit 13.

Subsequently, in step ST3, the discrimination unit 13 converts the position information (absolute coordinate position) of the bird input from the sound source separation and sound source localization unit 12 and the call of the bird, and the position information of each headset 20 into each other. On the basis of this, the type of bird is specified by the moss discriminating unit 13a from the song of the bird using the type database 13c. Then, the type discriminating unit 13a outputs the type information and the position information (absolute coordinate position) of the bird to the guide creating unit 14, and also transmits the bird call and the type information to the detailed judging unit 13b. It is sent to the detail discriminator 13b. Accordingly, in step ST4, the detailed discriminating unit 13b performs detailed discrimination of the bird using the detailed database 13d on the basis of the call and the type information of the bird, and guides the detailed information of the bird. Output to the creation unit 14.

Then, in step ST5, the guide creating unit 14 writes the position information (absolute coordinate position) of the bird type information, detailed information and position information (absolute coordinate position) from the discriminating unit 13 into Each head set 20 每 is converted into the relative coordinate position of the bird's position information based on the position information of each head set 20's GPS 22 and based on the above S ^ page information and detailed information. The commentary data on 20 is created and transmitted to the headset 20 by a transmitting unit (not shown).

As a result, in step ST6, each headset 20 received the explanation data created by the guide creation unit 14 by the transmission / reception unit 24, and performed noise cancellation by the noise cancellation circuit 25. Later, the user can use the commentary voice from the headphone 2 3 To provide. Therefore, the user can hear the comment based on the relative coordinate position of the bird to be observed from the user's position by the commentary sound heard from the headphone 23, and can easily find the bird, The bird's situation based on the call can be ascertained.

Thus, according to the audio guidance system 10 according to the embodiment of the present invention, the audio signals from the fixed microphones 11a, lib,... Sound source separation and sound source localization unit 12 performs sound source separation and sound source localization from the sound signal from movable microphone 21 of headset 20, and discriminator 13 recognizes the sound of birds The type and details are determined, and a description corresponding to the position of the headset 20 can be provided to the headset 20 of each user based on these. At this time, sound source separation and sound source localization are also performed with reference to the sound signals from the movable microphones 21 of each headset 20 so that more accurate sound source localization can be performed with respect to the position of each user. You.

In the embodiment described above, the voice guidance system based on the singing of a bird has been described. However, it is apparent that the present invention is not limited to this, and can be applied to a voice guiding system based on the singing of an animal other than a bird. Industrial applicability

As described above, according to the present invention, each fixed microphone of the microphone array and the movable microphone of each headset collect the sounds of animals and birds in the premises such as a natural park and the like, and , Sound source separation • The sound source localization unit separates the sound into the absolute coordinate position of each sound source and the sound data for each sound source. Then, the discriminating unit specifies the type of animal or the like that squeaks for each sound source from the sound source for each sound source separated by the sound source separation unit and the sound source localization unit. According to the type of the headset, explanation data is created corresponding to the position information of each headset, and the output unit outputs the explanation data to the headphone of the headset. This allows the user to obtain a commentary on the animal or bird that made the squeal by listening to the commentary flowing from the headphone of the headset carried by the user. The animal, bird, or the like can be easily found based on the coordinate position.

Therefore, according to the present invention, there is provided a voice guidance system with an extremely excellent cry which provides a user with an explanation of the animal or the bird based on the cry of the animal or the bird.

Claims

The scope of the claims

1. A microphone array consisting of a plurality of fixed microphones arranged as appropriate on the premises of a facility such as a natural park;

At least one headset consisting of the GFS, mobile microphone and headphone carried by the facility user;

Sound source separation and sound source localization unit for performing sound source localization and sound data separation for each sound source from sound signals from each fixed microphone and each headset movable microphone of the microphone array,

A discriminating unit that specifies a pest page of an animal or the like that has squeaked from acoustic data for each sound source separated by the sound source separation and sound source localization unit;

The absolute coordinate position of the sound source localized by the sound source separation / localization unit is converted into a relative coordinate position based on the positional information of each headset by GPS, and the type is determined by the type of animal or the like specified by the determination unit. And a guide creation unit for creating commentary data on the headset and outputting the data to the headphone of the headset.

2. The sound source separation / sound source localization unit includes a receiving unit that receives position information by GPS from each headset and an acoustic signal by a movable microphone,

The guide creation unit includes a transmission unit that transmits commentary data,

The headsets each include a transmitting / receiving unit that transmits position information by GFS and an acoustic signal by a movable microphone to the receiving unit, and receives commentary data from the transmitting unit. The voice guidance system described in 1 above.

3. The sound source separation and sound source localization unit performs sound source separation based on acoustic signals from the fixed microphones and the movable microphones by mathematical subtraction using a general micro phone array. 3. The voice guidance system according to claim 1, wherein

4. The voice guidance system according to any one of claims 1 to 3, wherein the sound source separation-sound source localization unit performs sound source separation using a direction pass filter.

5. The voice guidance system according to any one of claims 1 to 4, wherein the sound source separation / sound source localization unit further performs sound source separation by independent component analysis.

6. The method according to claim 1, wherein the discriminating unit refers to a type database having sounds of various animals and the like, and specifies the type of the animals and the like from the sounds of the animals and the like. Voice guide by any of the calls

7. The discriminating unit performs a detailed discrimination such as a sex discrimination of an animal or the like and a situation discrimination based on a kind of an animal or the like and further with reference to various detailed databases. The voice guidance system according to any one of (1) and (6).