JP2000207170A - Device and method for processing information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To present information easy to comprehend by users and to exchange additional information or high-accuracy information with a few errors between devices. SOLUTION: A CPU 21 generates an output audio signal by adding the additional information to an audio signal so as not to affect the hearing of voices due to that audio signal. The additional information is information related with the audio signal, for example. The voice composed of this output audio signal is outputted from a loudspeaker 12 towards an opposite side device. The opposite side device obtains an input audio signal by fetching that voice from a microphone 13 and supplies this audio signal to the CPU 21. The CPU 21 extracts the additional information from that input audio signal and performs processing, based on that additional information. For example, a phrase included in a natural language expressed by the voice is displayed on a display part 15. Furthermore, for example, voice recognizing processing is performed while limiting the object area of the natural language expressed by the voice with the additional information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an information processing apparatus and an information processing method capable of communicating information by voice. Specifically, by adding additional information to the audio signal in a manner that does not affect the listening of the audio by the audio signal, generating an output audio signal, converting the output audio signal into audio and outputting the audio signal, The present invention relates to an information processing apparatus or the like which can provide information that can be easily understood by voice, and realizes exchange of additional information and highly accurate information with little error between apparatuses.

[0002]

2. Description of the Related Art As a communication method between a plurality of information processing apparatuses, there is a method of connecting the apparatuses by using a cable or the like, or a method of performing wireless communication using infrared rays or radio waves. Generally, two devices are connected one-to-one, but communication between three or more devices can be performed by using a special device such as a hub on the way.

[0003] In input from a user to an information processing apparatus, information can be transmitted through a natural language by using a voice recognition technique. Further, by performing syntax analysis and semantic analysis of the natural language on the receiving side, various services can be provided to the user.

[0004] Outputs to a wide range of users include:
Audio output via speakers is widely used. Because the sound propagates over a wide area, and the necessary devices such as speakers can be realized at relatively low cost, for example,
It is implemented in the premises announcement system at stations and stores.

[0005]

However, in a method of connecting apparatuses by wire such as a cable, a terminal for connecting a cable to the apparatus is required, and a cable for connection is required. Further, in order to perform communication between three or more devices, a special device such as a hub is additionally required. As described above, prior preparation is required to perform communication, and there is a problem that communication cannot be performed immediately.

Further, in the method using infrared rays or radio waves, it is necessary to provide a special mechanism only for performing communication, and there is a problem that miniaturization and cost reduction of the apparatus are hindered.

[0007] In general, the user does not know what data is exchanged between devices. In order for the user to know this, there is a problem that means other than the communication path, such as screen display and audio output, must be used.

[0008] Further, when voice input is used as an input interface to an information processing apparatus, it is not possible to completely and correctly recognize a user's utterance with the current technology. In particular, since the number of recognized vocabularies greatly affects recognition performance, it is necessary to limit the number of vocabularies as much as possible in order to perform highly accurate recognition, and it is necessary to take measures such as narrowing down the recognition targets. Further, in the current speech recognition, information such as emotions inherent in the utterance of the user can hardly be obtained. Furthermore, it is very difficult for syntactic analysis and semantic analysis to derive a statement that completely reflects the input intention of the speaker.

[0009] Further, in an information processing apparatus provided with voice input / output, information can be exchanged interactively.
Information can be exchanged between devices by exchanging inquiries with each other, but no information is exchanged if the result of the inquiry is unknown. For this reason, in order to make an inquiry again, the user needs to input the voice again, and there is a possibility that a recognition error or the like may occur, which causes a problem that the user is burdened.

When voice output is used as means for presenting information to a user, for example, in the current announcement, no information is transmitted to a person who cannot hear. Further, since the language is expressed in a natural language, information is not transmitted even when there is a difference in a language used by a user. In addition, in-store announcements made by voice often make it difficult for the user to be familiar with the geography of the place, and it is difficult to understand places expressed in words.

Therefore, an object of the present invention is to provide an information processing apparatus and an information processing method that can solve the above-mentioned problems.

[0012]

According to a first aspect of the present invention, there is provided an information processing apparatus comprising: an audio signal generating means for generating an audio signal; an additional information generating means for generating additional information; And an audio output unit that outputs an audio signal based on the output audio signal by adding additional information in a manner that does not affect the listening of the audio signal.

An information processing apparatus according to a seventeenth aspect of the present invention
Voice input means for inputting voice and obtaining an input voice signal corresponding to the voice, additional information extracting means for extracting additional information added to the input voice signal, and information for processing using the extracted additional information Processing means.

The information processing apparatus according to claim 38 is
Audio signal generating means for generating an audio signal, additional information generating means for generating additional information, and information for obtaining an output audio signal by adding additional information to the audio signal in a manner that does not affect the listening of the audio by the audio signal Addition means, audio output means for outputting audio based on the output audio signal, audio input means for inputting audio to obtain an input audio signal corresponding to the audio, and addition for extracting additional information added to the input audio signal It is provided with an information extracting means and an information processing means for performing a process using the extracted additional information.

[0015] In the present invention, voice communication becomes possible. The voice in this case is an output voice signal obtained by adding additional information to the original voice signal in a manner that does not affect the listening of the voice by the voice signal. The additional information is, for example, information related to the original audio signal. For example, the addition of the additional information is performed by removing an attack portion in the audio signal, forming an instantaneous interruption section in a wide band portion, and using the instantaneous interruption section. Further, for example, the additional information is added to the audio signal as a spread spectrum signal. Thus, it is possible to present information that is easy for the user to understand by voice, and it is possible to realize exchange of additional information and highly accurate information with little error between devices.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an overview of an information processing apparatus 10 according to the first embodiment. A main body 11 of the information processing apparatus 10 is provided with a speaker 12 for outputting sound and a microphone 13 for inputting sound. The main body 11 has a talk switch 14 operated when inputting voice through the microphone 13.
Is provided. In this case, when the user operates the talk switch 14, voice input from the microphone 13 becomes possible.

The main body 11 is provided at its center with a display section 15 for displaying a GUI (Graphical User Interface) of the program. Further, on the surface of the display unit 15, a so-called touch panel (touch tablet) 16 that outputs a signal corresponding to the designated position when the user makes contact with the touch pen 1 or a finger or the like is arranged. ing.

Here, the touch panel 16 is made of a transparent material such as glass or resin. Therefore, the user can see the image displayed on the display unit 15 through the touch panel 16. Further, the user can use the touch pen 17 to input predetermined characters on the touch panel 16, select or execute a predetermined object (icon) displayed on the display unit 15, and the like.

FIG. 2 shows a circuit configuration of the information processing apparatus 10. The internal bus 20 has a CPU (central process
ng unit) 21, ROM (read only memory) 22, R
AM (random access memory) 23, display control unit 24,
The input interface 25 and the voice synthesizer 26 are connected to each other. Thus, each unit can transmit and receive data via the internal bus 20. CPU 21
Executes various processes according to programs or various data stored in the ROM 22 or the RAM 23.

The display control section 24 generates data of an image to be displayed on the display section 15 in accordance with the information supplied from the CPU 21 and causes the display section 15 to display the image. The input detection unit 27 includes the touch panel 16 and the talk switch 14.
Is detected, and a corresponding operation signal is supplied to the input interface 25. A / D converter 2
Reference numeral 8 denotes an audio signal output from the microphone 13;
An analog signal is converted into a digital signal and supplied to the input interface 25.

The input interface 25 receives the audio signal supplied from the A / D conversion section 28 or the operation signal supplied from the input detection section 27 and supplies it to the CPU 21. When an audio signal is input through the microphone 13, the A / D converter 28, and the input interface 25, the CPU 21
With reference to the data stored in the AM 23, a process of extracting additional information added to the audio signal is executed.

The voice synthesizer 26 generates a synthesized voice based on parameters and text data necessary for voice synthesis supplied from the CPU 21 and outputs the synthesized voice via the speaker 12. The voice synthesizing unit 26 is also used when reproducing voice recorded in the RAM 23 via the microphone 13. Further, when additional information to be transmitted to another device is supplied from the CPU 21, the voice synthesis unit 26 performs a process of adding the additional information to the synthesized voice or the recorded voice, and Output.

FIG. 3 shows a functional block diagram of the internal operation of the CPU 21 described above. The information processing section 30 performs various types of information processing. In the present embodiment, the information processing section 30 performs processing as an application program for managing a personal schedule.

The information adding section 31 generates an output audio signal SAout by adding specific additional information IFa provided from the information processing section 30 to the audio signal SAa provided from the information processing section 30. Here, the addition of the additional information IFa to the audio signal SAa is performed in a manner that does not affect the listening of the audio by the audio signal SAa.

For example, the additional processing is performed using a method described in Japanese Patent Application Laid-Open No. 10-162501. In this case, the information adding unit 31 performs the following additional processing. First, a sudden rising and large amplitude portion of the audio signal SAa is detected as an attack, and the audio signal Sa is detected.
A spectrum analysis is performed on a section of Aa having a predetermined length. Then, while removing the attack portion in the audio signal SAa, an instantaneous interruption section is formed in a wide band portion,
The output audio signal SAout is generated by adding the additional information IFa using the cutoff section. The attack part has a large effect on sound quality. Also, the wider the bandwidth, the harder it is to hear the click sound. Therefore, by performing the additional processing as described above, the click sound due to the instantaneous interruption can be made almost inaudible, and the sound signal SAa
Can be added to the additional information IFa.

Further, for example, when a certain loud voice signal is present in a human auditory characteristic, a low-level signal near the frequency cannot be heard or is very hard to hear. Processing may be performed. Further, for example, the additional information may be superimposed on the audio signal by the spread spectrum which has recently attracted attention. In this case, the additional information IFa
Is added to the audio signal SAa as a spread spectrum signal.

[0027] The information extraction unit 32 receives the input audio signal SAin.
A process of extracting additional information IFb from (corresponding to the above-described output audio signal SAout) is performed. The processing performed by the information extracting unit 32 depends on the processing method performed by the information adding unit 31. Here, additional information may not be inserted.

The information processing section 30 is a section for performing actual application processing.
Aa and the additional information IFa.
Supplies the additional information IFb or the audio signal SAb.

Next, the flow of processing performed in the functional block diagram shown in FIG. 3 will be described by taking information processing for schedule management as an example. FIGS. 4 and 5 show the flow of the processing, and show a case where voice (information) is transmitted to the outside and a case where voice (information) is input from the outside. In the following description, processing for managing an individual's schedule is used. However, the same procedure can be performed in other application examples.

The sound output operation will be described with reference to FIG. First, in step S40, the user selects information to be transmitted using input interfaces such as the display unit 15, the touch panel 16, and the touch pen 17. For example, the object “5/5 15:00 meeting” representing one item of the schedule on the display unit 15 is selected.
Then, in step S41, the information processing unit 30 generates a natural sentence (utterance sentence) that expresses the item selected by the user, for example, "a meeting is scheduled to be held from 15:00 on May 5". Further, additional information of a data expression that can be identified inside the system is generated.

In the following step S42, the speech synthesizer 26 generates a synthesized speech signal SAa for reading out the utterance sentence generated in step S41. Further, in step S43, the information adding unit 31 uses the voice synthesizing unit 26.
The additional information IFa generated in step S41 is added to the synthesized voice signal SAa generated in step S41. At this time, the additional processing is performed in a manner that does not affect the listening of the voice by the synthesized voice signal SAa, as described above. That is, the addition processing is performed by a method in which the change of the audio signal SAa due to the addition of the additional information IFa is indistinguishable or difficult to recognize by human hearing.

Finally, in step S44, the output audio signal SAout including the additional information IFa generated in step S43 is supplied to the speaker 12, and
Then, a sound based on the output sound signal SAout is output. The voice output in this manner is such that a human listening to it can understand the utterance sentence described above in natural language.
Further, in another information processing apparatus 10, by extracting the additional information IFb (same as IFa) from the input audio signal SAin relating to the audio, it is possible to understand the utterance sentence described above.

Next, the voice input operation will be described with reference to FIG. First, in step S50, an external voice is taken in by the microphone 13 to obtain an input voice signal SAin. Incorporation of the external voice is performed by, for example, signaling that the talk switch 14 is pressed by the user as a start signal, taking a period during which the talk switch 14 is being pressed as a capturing section, or detecting a voice of a certain level or more. Do.

In a succeeding step S51, the information extracting section 32 adds the additional information I included in the input audio signal SAin.
An Fb extraction process is performed. As a result of the processing, if the additional information IFb is not extracted, the processing ends. Additional information I
When Fb is extracted, the process proceeds to step S52, and the information processing unit 30 performs a process based on the additional information IFb. For example, when the sound described with reference to FIG.
In "in", data expressing "5/5 15:00 conference" is inserted as additional information IFb, and by extracting this additional information IFb, a process for adding / updating information is performed. Processing such as inquiring of the user for data update permission is performed.

Next, referring to FIG. 6, first information processing apparatus 10 (apparatus A) and second information processing apparatus 10 (apparatus B)
The operation of intercommunication between the two will be described. The device A is a device that makes an inquiry about the schedule, and the device B is a device that responds to the inquiry.

First, in the device A, the user Ua inputs a command for inquiring "what time is the meeting tomorrow". The input of this command is performed by the display unit 15 and the touch panel 1
6 and the touch pen 17. When the device A has a voice recognition function, the above-described command may be input by voice from the microphone 13. In step S60, the device A uses the user Ua
, An output audio signal SAout including the additional information IFa of "What time is the meeting tomorrow?" Is generated in the same procedure as the operation shown in the example of FIG. 4 described above, and the output audio signal SAout is output. Is output to the device B side. User Ua of device A and user Ub of device B
By listening to the audio signal, the user can clearly understand the processing that the device A intends to perform.

In the device B, in step S61, the sound output from the device A is taken in from the microphone 13, and processed by the same operation as the example shown in FIG.
"What time is the meeting tomorrow?"
Has been inserted as additional information IFb, and the apparatus B extracts the additional information IFb of this instruction in step S62. Then, in the device B, in the subsequent step S63, the additional information I
Information processing based on Fb, that is, information search for the time of the meeting tomorrow, is performed.
From 5 o'clock.

The result of the inquiry obtained in step S63 is automatically selected as it is as information to be transmitted in step S40 in FIG.
Alternatively, the apparatus B may request permission of the user Ub, and may be selected only when the permission is granted. In the device B, when the information to be transmitted is determined, in the subsequent step S64, the output audio signal SAout is generated by the same operation as in FIG. FIG. 6 shows, as an example, a case in which a response sentence “I do not understand” or “Because 15:00” is generated in S64. Here, an output audio signal SAout including not only a simple audio signal but also additional information IFa expressing a clear schedule is generated. In step S65, the audio based on the output audio signal SAout is output to the device A side.

In the device A, in step S66, the sound output from the device B is taken in from the microphone 13, and in step 67, the input sound signal SA
The additional information IFb expressing the schedule included in “in” is extracted. Then, if the additional information IFb is the information “from 15:00”, the process proceeds to step S68, where the addition / update to the schedule information of the device B is performed automatically or under the permission of the user Ua. I do. Through the interactive exchange between the device A and the device B as described above, it is possible to realize a process of taking in the information originally stored in the device B into the device A in response to an inquiry from the device A.

Further, if the extracted additional information is the information "I do not understand", the process proceeds to step S69, and the output audio signal SAout of the inquiry generated by the device A or the information indicating that the inquiry is unresolved is output. Information required to generate the data in the RAM in the device A
23. The stored information can be viewed by the user Ua via the display unit 15 and the touch panel 1
By selecting the information via the control unit 6, the audio based on the output audio signal SAout is output again. In this way, the inquiries that are not resolved in the dialogue between the devices are stored in the devices, and the voice relating to the inquiries can be re-output at any time thereafter, thereby reducing the user's work. It is possible to do.

Although the one-to-one mutual communication operation has been described with reference to FIG. 6, since voice is used as an information transmission medium,
One-to-many communication can be performed simultaneously. This allows
As in the examples given in FIGS. 4 and 5, the schedule information “5/5 15:00 meeting” issued from one information processing device 10 can be transmitted to a plurality of information processing devices 10 at the same time. Becomes

Further, the information processing apparatus 1 which outputs a sound by the sound signal as additional information to be added to the sound signal.
An identifier of 0 may be included. As a result, the information processing apparatus 10 incorporating the voice from the microphone 13
Can know which information processing apparatus 10 has output the sound, and can perform the processing in the information processing unit 30 in accordance with the information. For example, in the above-described example, if the schedule update information is from the reliable information processing apparatus 10, the information is automatically updated. Otherwise, the user is asked to confirm permission. Also, in the communication between the plurality of devices described above, each device can know from which device the sound is.

Further, interactive data transmission as exemplified in FIG. 6 is also possible. When such a dialogue is performed between a plurality of devices, it is possible that a plurality of devices output voice at the same time, so that the identifier of the device that grants the right to permit the next utterance is included as additional information. I do.
The device which has taken in the voice does not output the voice if the identifier included in the input voice signal SAin is different from its own identifier, and a plurality of devices can output voice simultaneously using this additional information. Can be prevented.

It is also possible to exchange memos by voice between a plurality of information processing apparatuses 10 to which the present invention is applied. In this case, the audio signal representing the note
By adding some keywords to which the contents of the memo belong as additional information, the receiving device can know the area to which the memo expressed by the voice belongs, and the receiving device automatically sorts and arranges the memo. can do.

Further, according to the present invention, it is possible to perform long-distance or wide-range communication using audio media as a medium. In the present invention, since voice is used for information transmission, it is possible to provide a wide range of information via a medium using voice, for example, a television or a radio. For example, information on the product can be obtained from a CM inserted into a television or radio.

Next, a second embodiment of the present invention will be described.

Although the information processing apparatus 10 according to the first embodiment does not have a voice recognition function,
The information processing apparatus 10 according to the embodiment has a voice recognition function. The circuit configuration of the information processing apparatus 10 according to the second embodiment is similar to that of the information processing apparatus 10 according to the first embodiment (see FIG. 2). Microphone 13, A / D converter 28 and input interface 2
When an audio signal is input via the input audio signal SAin, a process for extracting the additional information IFb from the input audio signal SAin is performed in the same manner as in the first embodiment.
Alternatively, the speech recognition processing is performed on the speech signal SAb with reference to speech learning data and dictionary information stored in the RAM 23.

In the current speech recognition technology, the recognition rate is 100
Instead of%, it often contains errors. Furthermore, it is very difficult for the system to extract subtle intentions and emotions contained in the voice. For example, the two sentences "Tomorrow is a meeting from 5 o'clock" and "Is tomorrow a meeting from 5 o'clock" have very different meanings depending on whether or not there is a "?" At the end. Is only a difference of one sound, and recognition errors are likely to occur.

Therefore, in the information processing apparatus 10 having a voice recognition function, additional information IFa for determining an intention such as “inquiry” or “question” is added to the voice signal SAa. Referring to FIG. 6 described above for explaining the intercommunication operation between the devices, the "inquiry" is added to the audio signal SAa from the device A to the device B as the additional information IFa.
Is added to the information processing device 10 on the receiving side, even if the "ka" at the end is overlooked during speech recognition, it is not interpreted as the intent of "determination", The interpretation / processing can be executed as a sentence.

Further, the information processing apparatus 1
When using when inputting information to 0, use the voice of the user,
Between devices, a recorded voice of the user or a synthesized voice generated in the device is used. Since these voice signals have different voice characteristics, different voice learning data are required in the ROM 22 or the RAM 23 for accurate voice recognition, and a large storage capacity is required. Therefore, speech recognition is used in the communication between the human and the information processing apparatus 10, and in the communication between the apparatuses, all information necessary for processing is given as additional information, or in order to avoid or reduce occurrence of recognition error. , Such as a keyword indicating a target area including an intention or a sentence, or information specifying a dictionary required for recognition.

For example, the voice output from the information processing device is related to "travel" and "inquiry"
In this case, the keyword “travel” and the intention information “inquiry” are inserted into the audio signal as additional information. In the information processing apparatus 10 which takes in the voice by the voice signal, "travel" and "inquiry" are first extracted as additional information, and then, the dictionary information for voice recognition suitable for "travel" and "inquiry" To execute voice recognition. As a result, the target area is limited, and speech recognition can be performed with higher accuracy.

Next, a third embodiment of the present invention will be described.

In the first and second embodiments described above, the present invention is applied to a personal information management device. However, the present invention can be applied to a robot.

The robot is provided with a voice recognition mechanism and a voice synthesis mechanism, and can interact with each other using a natural language. In utterances from humans to robots, mutual recognition is attempted by using voice recognition technology. However, the dialogue between robots is
Although it appears to be performed in a natural language, in reality, information is exchanged based on additional information added to the audio signal. This makes it possible to transmit highly reliable information between the robots while avoiding errors in speech recognition, and also enables the human observer to hear the contents as synthesized speech, thereby improving affinity and reliability. High communication can be realized.

In the industrial world, an event called RoboCup, in which teams are formed by a plurality of robots and a soccer game is played by two teams, is held every year. Here, a soccer game is played by various types of robots, but a spectator who observes the game does not know what the robot thinks and exchanges what information. Although various types of information are exchanged between robots, it cannot be known from the audience.

FIG. 7 shows an overview of a soccer game played by a robot to which the present invention is applied. A goal 70 is provided in a field 70 for playing soccer, and robots 73, 74, 75 play around a ball 72. FIG. 7 shows only a part of the field 70, and there are robots and goals not shown in FIG.

The robots 73, 74, and 75 are provided with a voice input / output unit. By exchanging information by mutually uttering voices, the robots are thinking what and exchanging what kind of information. Or, the audience 76 is made to know. In FIG. 7, the utterance “run before the goal” is made from the robot 73 to the robot 74, and the audience 7
6 can be known by listening to it as voice. Further, additional information IFa is added to an output audio signal SAout for outputting an audio generated by the robot 73,
Information transmission from the robot 73 to the robot 74 can be performed with high accuracy.

For example, as an additional information IFa, an identifier representing the uttered robot 73 and an instruction target are added to an output audio signal SAout for outputting a voice “run ahead of the goal” from the robot 73 to the robot 74. By adding an identifier representing the robot 74, its specific command, and the like, necessary information is transmitted between the robots. Furthermore, by giving the position coordinates as additional information IFa in response to the voice “before goal”,
It is also possible to exchange mutual positional information between the robots. Although sound has a certain degree of directivity, it has the property of propagating over a wide range. Therefore, it is possible for the robot 75 in FIG. 7 to take in voice and extract the additional information. However, a further distant robot that does not appear in FIG. 7 cannot sufficiently take in voice, and cannot exchange information by voice from the robot 73. This is the same property as a game played by a real human being, and it is possible to realize a more realistic game using a robot.

When it is considered that it is desirable not to utter a natural language, as in an animal-type robot, a bark may be used as a voice. In this case, it is very difficult for the current speech recognition technology to recognize a natural call and determine its content. By adding information indicating the intention and emotion of the call as additional information IFa to the output sound signal SAout for outputting the call, humans can only judge from the call and do not know as a clear language, but between robots. More clear communication can be achieved. As a result, it is possible to realize a robot that is close to the actual relationship between humans and animals, although some language is exchanged between animals, but humans do not know in detail.

A virtual zoo can be realized by applying the present invention. In this virtual zoo,
It is not an actual animal, but rather an animal cry and video, or a robot. Here, various kinds of animals are stored as information, and the user can see animals that cannot be normally seen.

Although humans can recognize the kind of animals to some extent from the sounds of animals, the current speech recognition mainly targets recognition of human language, and almost no consideration is given to sounds such as calls. . Therefore, it is difficult to identify the type from the cry. Also, the user does not know the type or name of an animal that is not very familiar.

Therefore, an output sound signal S for outputting a call
Information such as the type and name of the animal is added to Aout as additional information IFa. The user has the information processing device 10 according to the present invention, whereby the type and name of the animal added as additional information can be obtained from the sound of each animal. Further, the information processing apparatus can present more detailed information to the user based on the type and name. Thus, it is possible to realize a virtual zoo that presents the user with the type and name of the animal and more detailed information using the squeal of the animal as a key.

In the above description, the case where only the additional information is used for transmitting information between the robots has been described. However, it is needless to say that the voice recognition may be performed and the additional information may be used as an auxiliary.

Next, a fourth embodiment of the present invention will be described.

As an example of a system using the above-described information processing apparatus, an announcement system used for an announcement in a premises such as a station or a store can be considered. FIG. 8 shows an overview of the announcement system. The server 80 is for making announcements on the premises, and is connected to speakers 81 and the like installed at various places on the premises. User 8
Reference numeral 2 carries an information processing device 83 serving as a client, and listens to the sound output from the speaker 81 with its own ear, and the information processing device 83 takes in the sound.

The server 80 is the information processing device 1 shown in FIG.
0 is provided on the audio output side, and the output interface portion is connected to a plurality of speakers 81. The server 80 transmits, to the user 82 via the speaker 81, various information that needs to be provided over a wide area of the premises, such as notification of a lost child, arrival time of the next train, and product information such as a bargain sale, for example. Has functions. On the other hand, the information processing device 83 has a function on the voice input side of the information processing device 10 shown in FIG. 2, and has, for example, the appearance shown in FIG.

The server 80 is operated by an operator.
At any time, when an event to be announced occurs, a processing operation corresponding to the event is performed. The server 80 transmits the speaker 81
The voice in the natural language provided to the user 82 on the premises through is the voice of the operator, the voice recorded in advance, or the synthesized voice created as needed. However, for example, announcements about timetables in stations and announcements of store opening and closing, such as those that can be output automatically according to a predetermined schedule, are automatically output without the need for an operator. May be performed.

As the additional information to be added to the audio signal relating to the audio, for example, information expressing the content of the natural language can be used. Although it is very difficult for the information processing device 83 to recognize the voice of the announcement itself, the content of the announcement is added as additional information, and the content of the announcement is displayed using the display unit 15 of the information processing device 83. You can check.
Further, there is also an effect that it is possible to simultaneously make an announcement to a person who is deaf or hard to hear.

Also, in an announcement such as "... come to the 3rd floor elevator" which is given by the notification of a lost child, there is a problem that it takes time to find the place. In this case, position information of “in front of the third-floor elevator” can be provided as additional information to such an announcement. The information processing device 83 that has received the information displays the map information based on the position information on the display unit 15.
, It is possible to present more specific information to the user 82.

Further, a language different from the natural language expressing the announcement as the additional information, for example, a Japanese announcement is provided with information expressing a language such as English or French. As a result, the same information can be transmitted by voice to a foreigner who does not know the language of the country. The user 82 who cannot understand the meaning of the voice itself announced on the premises due to the language difference uses the information processing device 83 to add information expressing the translation into the native language to the voice signal. Then, it can be understood as a native language through the display unit 15 or the speaker 12.

In the above embodiment, the present invention is applied to the portable information processing device 10, the robots 73 to 75, or the server 80 as shown in FIG. 1, but the present invention is applied to other devices. Of course, the same can be applied.

[0072]

According to the present invention, an output audio signal is generated by adding additional information to an audio signal in a manner that does not affect the listening of the audio by the audio signal, and the output audio signal is converted into audio. The information can be output, and the user can be presented with information that is easy to understand by voice, and exchange of additional information and highly accurate information with few errors can be realized between the devices. In addition, since voice is used as a communication medium, information can be exchanged between a plurality of nearby devices at the same time, instead of one-to-one communication with directivity. Further, by using audio media such as television, radio, and telephone as a communication path, long-distance communication and communication for a large number of people can be performed.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an overview of an information processing apparatus according to an embodiment.

FIG. 2 is a block diagram illustrating a circuit configuration of the information processing apparatus.

FIG. 3 is a functional block diagram showing an internal operation of a CPU.

FIG. 4 is a flowchart for explaining a voice output operation (transmission operation) of the information processing apparatus.

FIG. 5 is a flowchart illustrating a voice input operation (reception operation) of the information processing apparatus.

FIG. 6 is a diagram for describing an example of a mutual communication operation between information processing apparatuses.

FIG. 7 is a diagram showing an overview of a soccer game played by a robot.

FIG. 8 is a diagram showing an overview of an announcement system.

[Explanation of symbols]

10 ... information processing device, 11 ... body, 12 ...
Speaker, 13 ... Microphone, 14 ... Talk switch, 15 ... Display unit, 16 ... Touch panel, 20 ... Internal bus, 21 ... CPU, 22 ...
ROM, 23 RAM, 24 display control unit
25 input interface, 26 voice synthesizer, 27 input detector, 28 A / D converter,
30 ... information processing unit, 31 ... information adding unit, 32
..Information extraction section

Continuing from the front page (72) Inventor Hirofumi Tsunoda 6-35, Kita-Shinagawa, Shinagawa-ku, Tokyo Inside Sony Corporation (72) Inventor Yasuhiko Kato 6-35, Kita-Shinagawa, Shinagawa-ku, Tokyo Sony Stock In-house F term (reference) 9A001 BB04 DD13 EE07 HH17 HH18 HH19 JJ38 JZ76

Claims (40)

[Claims] 1. An audio signal generating means for generating an audio signal; an additional information generating means for generating additional information; and adding the additional information to the audio signal in a manner that does not affect the hearing of the audio by the audio signal. An information processing apparatus, comprising: an information adding unit that generates an output audio signal by using the output unit; and an audio output unit that outputs audio based on the output audio signal. 2. The information processing apparatus according to claim 1, wherein the additional information is information related to the audio signal. 3. The information processing apparatus according to claim 2, wherein the audio signal represents a predetermined natural language, and the additional information is information indicating a phrase included in the natural language. apparatus. 4. The apparatus according to claim 2, wherein the audio signal represents a predetermined natural language, and the additional information is information used for recognition or interpretation of the natural language. Information processing device. 5. The apparatus according to claim 4, wherein the additional information is information indicating a keyword related to the natural language.
An information processing apparatus according to claim 1. 6. The information processing apparatus according to claim 4, wherein the additional information is information indicating an intention or an emotion expressed in the natural language. 7. The information processing apparatus according to claim 4, wherein the additional information is information for identifying a target area including the natural language. 8. The information processing apparatus according to claim 2, wherein the audio signal represents a place in a natural language, and the additional information is position information indicating the place. 9. The information processing apparatus according to claim 2, wherein the audio signal represents a predetermined natural language, and the additional information is information in another language corresponding to the natural language. apparatus. 10. The audio signal according to claim 2, wherein the audio signal corresponds to a human or animal voice, and the additional information is information indicating an intention or an emotion expressed by the voice. Information processing device. 11. The information processing apparatus according to claim 2, wherein the audio signal corresponds to a sound of an animal, and the additional information is information indicating a type and a name of the animal. 12. The information processing apparatus according to claim 1, wherein the additional information is identification information indicating its own apparatus. 13. The information processing apparatus according to claim 1, wherein the additional information is information for identifying a device having a right to output the sound next. 14. An information adding means, comprising: an attack detection unit for detecting a sudden rising and large amplitude portion of an audio signal as an attack; and a spectrum analysis unit for performing spectrum analysis on a section of the audio signal having a predetermined length. An instantaneous interruption section forming section for removing the attack part in the audio signal from the output of the attack detection section and the output of the spectrum analysis section, and forming an instantaneous interruption section in a wide band portion; 2. An information adding unit for adding the additional information by using the determined instantaneous interruption section.
An information processing apparatus according to claim 1. 15. The information processing apparatus according to claim 1, wherein the information adding means adds the additional information to the audio signal as a spread spectrum signal. 16. A step of generating an audio signal; a step of generating additional information; and generating an output audio signal by adding the additional information to the audio signal in a manner that does not affect the listening of the audio by the audio signal. And a step of converting the output audio signal into audio and outputting the audio. 17. An audio input means for inputting audio and obtaining an input audio signal corresponding to the audio, additional information extracting means for extracting additional information added to the input audio signal, and the extracted additional information An information processing apparatus comprising: an information processing unit that performs a process using information. 18. The information processing apparatus according to claim 17, wherein the additional information is information related to the voice. 19. The information processing apparatus according to claim 18, wherein the voice represents a predetermined natural language, and the additional information is information indicating a phrase included in the natural language. . 20. The information processing apparatus according to claim 19, wherein the information processing means displays the phrase on a display unit using the additional information. 21. The information processing apparatus according to claim 18, wherein the voice represents a predetermined natural language, and the additional information is information necessary for recognition of the natural language. 22. The information processing apparatus according to claim 21, wherein the additional information is a keyword related to the natural language. 23. The information processing apparatus according to claim 21, wherein the additional information is information indicating an intention or an emotion expressed in the natural language. 24. The information processing apparatus according to claim 21, wherein the information processing means performs a voice recognition process for recognizing the natural language from the input voice signal. 25. The apparatus according to claim 2, wherein the additional information is information specifying a dictionary necessary for the voice recognition.
4. Information processing measures according to 4. 26. The information processing apparatus according to claim 24, wherein the additional information is information for identifying a target area including the natural language. 27. The information processing apparatus according to claim 18, wherein the voice is a representation of a place in a natural language, and the additional information is position information indicating the place. 28. The information processing apparatus according to claim 27, wherein the information processing means displays the location on a display unit using the additional information. 29. The information processing apparatus according to claim 18, wherein the voice represents a predetermined natural language, and the additional information is information of another language corresponding to the natural language. . 30. The information processing apparatus according to claim 29, wherein the information processing means displays the foreign language on a display unit using the additional information. 31. The voice is a human or animal voice,
19. The information processing apparatus according to claim 18, wherein the additional information is information indicating an intention or an emotion expressed by the voice. 32. The information processing apparatus according to claim 18, wherein the voice is a sound of an animal, and the additional information is information indicating a type and a name of the animal. 33. The information processing apparatus according to claim 32, wherein the information processing means displays the type and name of the animal on a display unit using the additional information. 34. The information processing apparatus according to claim 17, wherein the additional information is information for identifying a device that has output the sound. 35. The information processing apparatus according to claim 17, wherein the additional information is information for identifying a device having a right to output the sound next. 36. The information processing apparatus according to claim 17, wherein said information processing means is a control means for controlling an operation of a robot imitating a living thing. 37. A step of converting an input voice to obtain an input voice signal, a step of extracting additional information added to the input voice signal, and performing a process using the extracted additional information. And an information processing method. 38. An audio signal generating means for generating an audio signal, additional information generating means for generating additional information, and adding the additional information to the audio signal in a manner that does not affect the listening of the audio by the audio signal. Information adding means for obtaining an output audio signal by performing the above operation; audio output means for outputting an audio based on the output audio signal; audio input means for inputting the audio and obtaining an input audio signal corresponding to the audio; An information processing apparatus, comprising: an additional information extracting unit that extracts additional information added to a file; and an information processing unit that performs a process using the extracted additional information. 39. The information processing apparatus according to claim 38, wherein said information processing means is a control means for controlling an operation of a robot imitating a living thing. 40. When the input voice signal including the information for resolving the query is not obtained from the voice input means after outputting the voice of the output voice signal relating to the query from the voice output means, the information of the query is provided. Information storage means for storing the unresolved inquiry information as unresolved inquiry information, and based on the unresolved inquiry information stored in the information storage means, from the voice output means at an arbitrary timing. 39. The information processing apparatus according to claim 38, further comprising: re-output control means for outputting a sound based on the output sound signal according to the inquiry.