JPH06208389A - Method and device for information processing - Google Patents

Abstract

PURPOSE:To realize natural interaction between a device and a user in the device which retrieves information performing voice conversation, predicts a next step conversation by a user, and selects an object to be recognized. CONSTITUTION:This processing method is performed as shown in the flow chart. That is, input information is sent to a voice recognizing section and voice recognition is performed 202. The recognized result is sent to a conversation control section 203, it is judged whether the recognized result satisfies retrieving conditions or not 204, if the result satisfies the conditions, indication for retrieving is issued 206, and if not, indication for continuing conversation is issued 205. when retrieving conditions are arranged, information is retrieved from a data base in an information retrieving section 207, answering of conversation is generated based on an output information from the conversation control section and the information retrieving section in a conversation answering generation section 208, and it is outputted to the voice output section or a display device. A next step conversation is predicted considering conditions of retrieving information and conversation 209, when the voice is outputted, after a predicted object to be recognized is generated in a generation section for the object to be recognized 210, procedure is turned to the original and the voice input of a next conversation is expected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing method and apparatus for interacting with a user by inputting a language by voice.

[0002]

2. Description of the Related Art Voice is the most naturally used of information exchange means between humans. On the other hand, due to the dramatic progress of computers, computers have advanced to handle various information as well as numerical calculation. Therefore,
There is a demand to use voice as a means of exchanging information between humans and computers.

The conventional voice information search apparatus does not dynamically change the information for performing voice recognition on the search item, or even if there is a change, it follows a certain flow of dialogue in advance. Only the changes were made, and it was possible to use only the words and sentences registered in advance, and it was not possible to perform a natural voice search.

Conventionally, all user language input has been performed by voice input.

[0005]

The conventional voice information retrieval apparatus has a problem that the voice retrieval can be performed only by using the words or sentences registered in advance. Further, in a state of a certain dialogue, since only predetermined certain pre-registered dialogue contents can be recognized, there is a problem that natural dialogue cannot be sequentially performed. For this reason,
There was a problem that all search items on the database could not be searched naturally.

[0006] Further, in general, when a dialogue is conducted naturally,
There are inputs that can occur anywhere in the dialogue. For example, in a dialog for searching travel information, there are meta questions such as "What kind of items can you ask?" And very global questions such as "I want to know a golf course in Tokyo." On the other hand, as the dialogue progresses, questions related to detailed contents, such as
"I would like to know the telephone number of Yumoto Hot Spring in Hakone." Or "What is the address of the hot spring in Yoshii Town, Gunma Prefecture?" By recognizing static speech recognition information for accepting input that can be uttered anywhere in this dialogue and voice recognition, and input that changes dynamically as the dialogue progresses Enlargement, low recognition performance, and complicated control are problems.

If the user's language input is only voice, or if there is a language in which voice recognition is not possible, there is a problem that the dialogue does not proceed.

[0008]

In order to solve the above problems, the present invention has a dictionary for inputting voice and recognizing voice, recognizing the input voice using the dictionary, and the result of the recognition. And an information processing method and apparatus for changing the dictionary according to the search result.

In order to solve the above problems, preferably, the change of the dictionary is updating.

In order to solve the above-mentioned problems, preferably the change of the dictionary is addition.

In order to solve the above problems, preferably, the change of the dictionary is deletion.

In order to solve the above-mentioned problems, it is preferable to judge whether or not reading information is attached to the search result, and if not, the reading information is added.

In order to solve the above problems, preferably, the search result is displayed together with reading information of the search result.

In order to solve the above problem, preferably, a grammar according to the search result is selected, and the next utterance is predicted according to this grammar.

In order to solve the above problems, preferably, character information is received, and the character information is controlled so as to be processed together with the recognition result of the input voice.

[0016]

【Example】

(Embodiment 1) FIG. 1 is a block diagram showing the configuration of a voice conversation information retrieval apparatus in this embodiment.

In FIG. 1, 101 is a microphone for inputting voice, 102 is a speaker for outputting voice, 103 is an A / D, D / A converter for converting signals input / output from the microphone 101 and the speaker 102, 104. Is a display device capable of displaying images such as a CRT, 105 is a database,
Further, a ROM for storing a control program of a flow chart described later, (read only memory), 106 stores various data and is used as a working memory.
AM (random access memory), 107 is the ROM 10
5 is a CPU (central processing unit) that controls the entire apparatus based on the control program stored in FIG.

FIG. 2 shows a flowchart of voice recognition and dialogue processing in this embodiment.

FIG. 3 is a functional block diagram of the voice interactive information retrieval apparatus in this embodiment. The overall processing flow of this embodiment will be described with reference to the flowchart of FIG. 2 and the functional configuration diagram of FIG.

First, an example of dialogue for explaining the processing will be shown. (Usr: Generation of user, Sys: Speaking of voice dialogue system in this embodiment) Usr: "I want to know the park in Tokyo." Sys: "10 in Chiyoda Ward, 5 in Setagaya Ward, and so on." Usr: "In Setagaya-ku." Sys: "Kinuta Park, Ashihana Park, ...." Usr: "Tell me the phone number of Kinuta Park." Sys: "03- ×××-×××××" Usr: "Please tell us about the shrines in Setagaya Ward." ...
(1) Sys: “Yawata Shrine, Karasuyama Shrine, ...” Usr: “The address of Karasuyama Shrine is.” Sys: “Setagaya-ku, Minamikarasuyama △ － △ － △.” Such a natural dialogue The possible processing will be described.

Voice input is performed using the microphone 101 (201), input information is sent to the voice recognition unit 312, and voice recognition is performed (202). The recognition result is sent to the dialogue management unit 302 (203), it is judged whether or not the recognition result satisfies the search condition (204), and if the condition is satisfied, a search instruction is issued (206). If not, an instruction is given to continue the dialogue to obtain the missing information (205). If the search conditions are satisfied, the information search unit 303 searches the database for information according to the search instruction (207). Based on the information output from the dialogue management unit 302 and the information retrieval unit 303, the dialogue response generation unit 320 generates a dialogue response (2
08), and outputs the generated response to the audio output unit 102 and the display device 104. The next utterance is predicted based on the retrieved information and the state of the dialogue (209), and the recognition target generation unit 320 generates a recognition target predicted to be uttered in the next utterance (210). When the recognition target is generated (210),
The process returns to 201 and waits for the next speech input.

Here, reading is previously attached to the search item of the database searched in step 207, and the ROM 105 is attached.
Stored in. In step 207, it is determined whether or not the reading is attached to the item retrieved, and if not, the reading is taken out from the dictionary in the ROM 105 and attached to the retrieval item. Also, the reading is attached to the recognition target generated in step 210, and the reading device 1 displays the reading information.
It is displayed on 04.

FIG. 3 illustrates the recognition object generator in detail. The recognition target generation units 304 to 311 surrounded by the dotted line in the figure. In general, when a dialogue is conducted naturally, there are inputs that can be spoken anywhere in the dialogue. For example, in a dialog for searching travel information, there are meta questions such as "What kind of items can you ask?" And very global questions such as "I want to know a golf course in Tokyo." On the other hand, as the dialogue progresses, questions related to detailed contents, such as
"I would like to know the telephone number of Yumoto Hot Spring in Hakone." Or "What is the address of the hot spring in Yoshii Town, Gunma Prefecture?" Accept input that can be spoken anywhere in this dialogue. Use the word dictionary and grammar for voice recognition as the static word dictionary unit 304 and the static grammar unit 305, and accept the input that dynamically changes as the dialogue progresses. A word dictionary for speech recognition is generated by the next utterance word dictionary generation unit 306 by the search content word dictionary generation unit 307, and the grammar is generated by the dynamic grammar unit 309 according to the content of the generated word dictionary. Using the dynamic grammar selection unit 310, the recognition target generation unit 311 creates the information together with the static grammar information of the static word dictionary unit 305 304.

That is, in this embodiment, information that can be a recognition target predicted to be uttered in the next utterance is given in the following 2
It has a means for holding it as one piece of recognition information.

(1) Static recognition information (word dictionary and grammar) for recognizing a sentence that can be input at any time regardless of the dialogue situation.

(2) Dynamic recognition information (word dictionary and grammar) in which the vocabulary and grammar to be recognized dynamically change according to the conversation situation.

The timing of rewriting the held information can be the following three cases.

(A) Immediately update the contents held in the same item.

(B) Even if a change is made, the contents held in the same item are not updated, but the information is newly added and held.

As shown in (c) and (b), while holding information that has once appeared in the dialogue, as the dialogue progresses, the held information for recognition is sequentially erased according to the conditions of the dialogue. To do.

A process for enabling the natural dialogue described above by using any of the above (a), (b), and (c) will be described. For example, the last utterance was "I want to know a park in Tokyo."
If the search result in “Inokashira Park, ..., Shiba Park, Yoyogi Park” is used, first, using this result, the park name of this search result is used as the recognition target vocabulary to be created in S210. Select a grammar that suits your vocabulary and create a dynamic grammar. By developing this dynamic grammar and constructing a dynamic network in the speech recognition unit, in the next dialogue,
Be able to recognize the sentence “Please give me the telephone number of Shiba Koen”.

When there are a large number of search results in response to the search command, the dialogue management unit does not output the search results,
Output "10 in Chiyoda-ku, 5 in Setagaya-ku, ...",
This city name is used as a recognition target word in the next dialogue, and a grammar suitable for these vocabularies is selected to create a dynamic grammar. By developing this dynamic grammar and constructing a dynamic network in the speech recognition unit, it becomes possible to recognize a sentence using the municipality name "In Setagaya Ward." In the next dialogue.

In other words, in the first example, the retrieval result is directly used as the next utterance prediction word, whereas in the second example, the retrieval result is so large that there are many retrieval items. The place name that guides the next utterance is used as the next utterance prediction word so that the narrowing down can be performed.

Further, the retrieval result is "park" or "place name", and a grammar suitable for each is selected to prepare for the next utterance.
In the case of a park, you can enter the location, size, direction, etc .. In the case of a place name, the park, facility,
Select a grammar that allows you to enter….

A dynamic network for recognizing the recognition target vocabulary and grammar determined from these search results, and the recognition target vocabulary and grammar of the municipality name for narrowing down by knowledge, both at the same time or only one of them. Is created. In addition, even if any of the holding information changing methods (a), (b), and (c) is used, recognition of these words is performed unless the area or item to be recognized is explicitly changed in the dialogue. It has a feature that can do.

In other words, since the information about the place name is not rewritten by the new information about the park, even if the center of the dialogue is the park, it becomes possible to recognize the place name that was involved in the previous dialogue. If all the information is stored, the recognition unit becomes huge. Therefore, by deleting (c), it is possible to recognize the information of past dialogue to some extent and avoid the problem of the recognition unit becoming huge. Is possible.

FIG. 4 is a diagram of the voice recognition unit, which is shown in FIG.
04 (405 to 412) is 105 in FIG. 1, and in FIG.
It hits 202. Reference numeral 403 is a recognition model (standard pattern) for recognition, and by using these, a network for recognition of 406 and 412 is realized inside the computer based on the character information of 405 and 411. This network does not have to be created in advance before starting recognition, and can be dynamically created according to voice input. 40
The voice waveform input from 1 is subjected to acoustic analysis in 402 and converted into acoustic parameters. This acoustic parameter and the route of 406 and 407 which is most likely to be the input voice on the network is determined (408) and is set as the first recognition result. The second input voice-like route on all networks is the second candidate, the third is the third candidate, .... In this example, there are three recognition networks, a static network 406 and a dynamic network 4 respectively.
12 consists of two things, hot spring related and park related. Therefore, the recognition result of the first place is the first place (413, 4) of each network.
14, 415), the result which seems to be the most input voice will be selected.

The static grammar 405 allows input that can occur anywhere in the interaction, in this example creating a static network 406. Dynamic network 412
Is created from the dynamic grammar 411. The grammar is in accordance with the input of the previous utterance 407 “Tell me about the park in Tokyo.” The dialogue search management unit 408 (103 in FIG. 1,
Hit 107. ), “Inokashira Park, Shiba Park, ..., Yoyogi Park,” 409, and is created by the grammar generation unit 410.

FIG. 5 shows an example in which search items and prediction items are output. 501: "Show me the golf course in Atsugi."
Is the search result of. Also,. 502 is the output of the name of the municipality requested as the next utterance as a condition for further reducing the number of search items because the number of search items is large in response to the input "Show me the golf course in Kanagawa." In the present invention, a more natural dialogue is realized by selecting a grammar suitable for each item based on the information of the search item of the database to be searched and the information of the next utterance item, and managing the information independently for each item. It is characterized by doing.

FIG. 6 shows an example in which new search items and predicted items are output after the next dialogue.
511 is the search result for "Show me the hot springs in Hakone."
Further, reference numeral 512 is the output of the name of the municipality requested as the next utterance as a condition for further reducing the number of search items because the number of search items is large in response to the input “Show hot springs in Kanagawa prefecture”.

When the recognition target switching (a) of the present invention is used, when 512 is output, the same place name 5 is output.
02 was rewritten immediately and read "Atsugi City, Yokosuka City, ..."
Becomes unrecognizable in the next utterance. That is, (a)
According to the method, it becomes possible to recognize 501, 511 and 512 in the next utterance. On the other hand, if (b) and (c) are used, 512 is added to 502, 501, 502, 511,
Recognition regarding 512 becomes possible. According to the method of (c), the information of 502 is deleted after the dialogue progresses for a while.

FIG. 7 is an example of a dynamic grammar created when the search item is a golf course. The search result is 501 in FIG.
At this time, the word dictionary 602 is created, and the grammar 602 suitable for the golf course is selected. As a result, the word grammar information for creating the recognition network 601 is obtained.

FIG. 8 is an example of a dynamic grammar created when the next speech prediction item is a place name. Figure 5 shows the next utterance prediction item
No. 502, the word dictionary 702 is created, and the grammar 702 whose next utterance prediction item is the place name is selected. As a result, 7
The word grammar information for creating the recognition network indicated by 01 is obtained.

FIG. 9 shows an example of a word dictionary and grammar for accepting input and voice recognition that can occur anywhere in this dialogue. Information such as 802 is stored in the static word dictionary portion 304 of FIG. 3, and 80 is stored in the static grammar portion 305.
The grammar like 3 is stored. At the time of recognition, 801
The recognition network is created and recognition is performed.

In 901 of FIG. 10, an example of a sentence that can be recognized by the word grammar information for creating the recognition network of FIG.
2 shows an example of a sentence that can be recognized by the word grammar information that creates the recognition network of FIG. 8, and 903 shows an example of a sentence that can be recognized by the word grammar information that creates the recognition network of FIG.

As described above, according to this embodiment, it is guaranteed that the information retrieval by voice input can be realized in a natural and easy-to-use form.

In the present embodiment, the holding of the information of the recognition item is described at the level of the word dictionary. However, in addition to this, the information is held in the state of the recognition network 406 or 412 shown in the fourth, and each item is held. It is also possible to hold and manage.
(2) Further, if the method described for holding information of recognition items is used, the recognition networks 406 and 412 can be made into one large recognition network by dynamically using all the held information. Further, it is also possible to select a number that is irrelevant to the number of items to be managed by a method suited to the arithmetic element of the recognition device.

(Embodiment 2) The present invention will be described in detail below with reference to the drawings.

FIG. 11 is a block diagram showing the basic configuration of an apparatus according to an embodiment of the present invention. The present embodiment is an embodiment of a voice dialog device that searches for a document along with a dialog.
In FIG. 11, 1 is a voice input unit that accepts the voice input of the user, 2 is a voice recognition unit that converts the voice input of the user into language information, and 3 is a character input unit that accepts the character input of the user.
Reference numeral 4 indicates the language information which is the recognition result of the voice recognition unit 2 if there is a user's voice input, and the character information accepted by the character input unit 3 if there is a user's character input, as input of the same language information. A dialogue processing unit 5 for performing a dialogue, a recognition range presenting unit 5 for presenting a range of languages converted by the voice recognition unit 2 to a user,
Reference numeral 6 denotes a dialogue output unit 7 for outputting information output from the dialogue processing unit 4 to the user, and a dialogue processing unit 7 for searching a document in response to a request from the dialogue processing unit 4.

FIG. 12 is a diagram showing a concrete system configuration of the embodiment of the present invention. Here, 21 is a control memory, which stores a control program according to a control procedure as shown in the flowchart of FIG. Reference numeral 22 is a central processing unit for making judgments / calculations and the like according to the control procedure stored in the control memory 21. A microphone 23 realizes the voice input unit 1 shown in FIG. A voice recognition device 24 realizes the voice recognition unit 2 shown in FIG. A keyboard 25 realizes the character input unit 3 shown in FIG. Two
A CD-ROM drive 6 holds a CD-ROM containing a document to be searched. Reference numeral 27 denotes a display, which realizes the recognition range presenting unit 5 and the dialogue output unit 6 shown in FIG. 28 is a bus.

The processing of this apparatus will be described below with reference to the flowchart shown in FIG. In this embodiment, a database search process is used as an example of the process performed by the dialogue processing unit 3.

First, in S1, the voice recognition unit 2 and the dialogue processing unit 4 are initialized. Then, the range of languages converted by the voice recognition unit 2 is passed to the recognition range presenting unit and presented to the user. In addition, a message prompting the user to input from this device is output to the dialogue output unit 6. Then, the process proceeds to S2. In S2, it is checked whether or not voice recognition by the voice recognition unit 2 is performed as a result of voice input to the voice input unit 1, and if there is a recognition result, the process proceeds to S4. If not, go to S3. In S3, it is checked whether or not there is a character input in the character input unit 3, and if there is an input, the process proceeds to S5. If not, the process returns to the beginning of S2. In S4, the speech recognition result is fetched into the dialogue processing unit 4 as character information. In step S5, the character input input to the character input unit is fetched into the dialogue processing unit 4. S
In 6, the dialog processing is performed by using the captured character information as the input of the user. In interactive processing, sentence analysis of the input sentence is performed,
Extract the user's intention and perform the corresponding processing. Here, the document is searched. Then, based on the search result, an output to the user is created and sent to the dialogue output unit 6. In S7,
Based on the result of the dialogue processing, it is determined whether or not the dialogue is to be ended. If the dialogue is to be ended, all processing is ended. When not ending, the process returns to the beginning of S2.

FIG. 14 shows an example of the recognition range in this embodiment. FIG. 15 shows an example of the dialogue between the user and the device. In this dialog example, the input of the user 1 is performed by voice input, and the input of the user 2 is performed by character input.

In the present embodiment, the language to be converted by the voice recognition means is limited to a specific range, and the recognition range presenting means for presenting the range to the user is provided.
It is also possible to eliminate the recognition range presenting unit 5 of the block diagram of FIG. 13 so that the recognition range is not presented in step S1 of the flowchart of FIG. 13, and a speech recognition apparatus that does not specify the language to be converted.

Further, the case where the voice input unit is separately provided has been described, but the present invention is not limited to this, and a voice recognition unit capable of direct voice input may be used without providing the voice input unit.

Although the case where the dialogue output section is provided has been described, the present invention is not limited to this, and the dialogue output section may not be provided when there is no dialogue output.

Further, although the case where the input to the interactive process and the output from the interactive process are all performed in the language has been described, the present invention is not limited to this, and command input or table output may be performed.

Further, although the case where the document is retrieved along with the interactive processing has been described, the present invention is not limited to this, and any processing carried out through the interactive such as guidance, education, and calculation may be performed. Further, only the dialogue may be performed without performing other processing.

Further, the case where the presentation of the recognition range and the interactive output are output to the same display has been described, but the present invention is not limited to this and may be output to different displays.

Also, the case has been described in which whether or not there is a voice input is determined based on the presence or absence of the result of voice recognition.
The present invention is not limited to this, and it may be determined whether or not there is an input to the voice input unit.

Further, although the case where the dialogue is started by the output of the message from the device to the user has been described, the present invention is not limited to this, and may be started by the input from the user to the device.

Further, although the case where the voice recognition means is realized by the voice recognition device has been described, the present invention is not limited to this, and any voice recognition means such as a software implemented on a computer may be used.

Although the case where the character input means is realized by the keyboard has been described, the present invention is not limited to this, and any means that can input a character code such as a pen input device or a touch panel may be used.

Further, the case where the recognition range presenting means is realized by a display has been described, but the present invention is not limited to this, and an arbitrary recognition range such as output by a printer or voice output by a voice synthesizer is possible. A presentation means is sufficient.

Although the case has been described where the document to be searched is a CD-ROM document in a CD-ROM drive, the present invention is not limited to this, and any document such as a document on a hard disk may be used.

Also, the case where the voice dialogue apparatus of the present invention is realized by using a general-purpose computer has been described, but the present invention is not limited to this, and a part or all of the processing according to the present invention uses dedicated hardware. May be realized.

Further, the case where the voice dialogue apparatus of the present invention is realized on one general-purpose computer has been described, but the present invention is not limited to this, and communication is performed between a plurality of general-purpose computers and dedicated hardware. May be realized.

[0068]

The present invention divides and holds two or more pieces of recognition information into static information and dynamic information, and the dynamic information is used as a word dictionary, grammar, or recognition network for each item to be recognized. By managing, the above-mentioned problems can be solved and a natural and easy-to-use spoken dialogue information retrieval device can be realized.

The present invention is a database in which reading information is preliminarily attached to search items of a database and items of predicted next utterance information, and dynamic and automatic reading information is added to search items of a database and items of predicted next utterance information. Means to attach,
This is a problem in the conventional voice information retrieval device by using the means for outputting the retrieval item of the database and the item of the next utterance information to the display device as the information with the reading attached to guide the predicted next utterance. Also, since the information for voice recognition is not attached to the search items, only the words or sentences registered in advance can be used to search by voice, and therefore all the search items in the database cannot be searched. It is necessary to pre-read all the search items in the database in advance in order to solve the problems and the search items in the database. If the database is changed, read the data. There is a problem that this information has to be incorporated into the voice recognition device in advance, and even if read information is incorporated into voice recognition in advance, for example, Since I do not know how to say difficult Kanji such as words, I can not input information after all, so I solved the problem that information search could not be done, and a natural and easy-to-use spoken dialogue information search device realizable.

According to the present invention, in the voice dialogue information retrieval apparatus, a word which is likely to be uttered in the next utterance or a means for selecting a word which is desired to be uttered, and a prediction word are selected in accordance with the situation of the dialogue and the inspection result. By using the selection means to select the appropriate grammar, it is possible to retrieve information by natural spoken dialogue.

As described above, the voice interactive apparatus of the present invention accepts the language information which is the voice recognition result of the voice input when the user inputs the voice, and the character input means when the user inputs the character. Since the dialogue information is processed by inputting the same character information as the input of the same language information, the effect that the dialogue can proceed normally by inputting the character even if the voice recognition fails is there.
In addition, even if the range of language input to be recognized is limited in order to improve the accuracy of voice recognition, input exceeding the limited range can be performed by character input, so the user's input is not limited and any language The effect is that you can enter. Further, since the range of languages to be recognized is presented to the user, there is an effect that it is possible to determine whether to use voice input or character input depending on the content that the user is trying to input. .

[Brief description of drawings]

FIG. 1 is a configuration diagram of this embodiment.

FIG. 2 is a flow chart of processing of this embodiment.

FIG. 3 is a diagram of a recognition target generation unit according to the present embodiment.

FIG. 4 is a diagram of a voice recognition unit according to the present embodiment.

FIG. 5 is an example of information on search items and prediction items according to the present embodiment.

FIG. 6 is an example of information of a newly presented search item and prediction item of the present embodiment.

FIG. 7 is an example of a dynamic grammar created when the search item of this embodiment is golf.

FIG. 8 is an example of a dynamic grammar created when the prediction item of this embodiment is a place name.

FIG. 9 is an example of a static grammar that enables voice input in any of the conversation situations of this embodiment.

FIG. 10 is an example of a sentence that can be recognized by each grammar of this embodiment.

FIG. 11 is a block diagram showing a basic configuration of a voice interaction device according to a second embodiment.

FIG. 12 is a diagram showing a specific system configuration of the second embodiment.

FIG. 13 is a flowchart showing the outline of the processing procedure of the second embodiment.

FIG. 14 is a diagram showing an example of a recognition range in the second embodiment.

FIG. 15 is a diagram showing an example of a dialogue in the second embodiment.

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Keiichi Sakai, 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc. (72) Minor Fujita, 3-30-2 Shimomaruko, Ota-ku, Tokyo (72) Inventor Takaya Ueda 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc.

Claims (16)

[Claims] 1. A voice input and a voice recognition dictionary are provided, the input voice is recognized using the dictionary, information is searched according to the recognition result, and the dictionary is changed according to the search result. An information processing method characterized by the above. 2. The information processing method according to claim 1, wherein the change of the dictionary is an update. 3. The information processing method according to claim 1, wherein the change of the dictionary is addition. 4. The information processing method according to claim 1, wherein the change of the dictionary is deletion. 5. The information processing method according to claim 1, wherein it is determined whether or not reading information is attached to the search result, and if not, the reading information is added. 6. The information processing method according to claim 1, wherein the search result is displayed together with reading information of the search result. 7. The information processing method according to claim 1, wherein a grammar according to the search result is selected, and the next utterance is predicted according to the grammar. 8. The information processing method according to claim 1, wherein the character information is received, and the character information is controlled so as to be processed together with the recognition result of the input voice. 9. An input unit for inputting voice, a dictionary for recognizing voice, a recognizing unit for recognizing the input voice using the dictionary, an information retrieving unit for retrieving information according to a result of the recognition, An information processing apparatus comprising: a changing unit that changes the dictionary according to a search result. 10. The information processing apparatus according to claim 9, wherein the change of the dictionary is an update. 11. The information processing apparatus according to claim 9, wherein the change of the dictionary is addition. 12. The information processing apparatus according to claim 9, wherein the change of the dictionary is deletion. 13. The information according to claim 9, further comprising: a determining unit that determines whether or not the reading result has reading information, and an adding unit that adds reading information if the reading result does not have reading information. Processing equipment. 14. The information processing apparatus according to claim 9, further comprising display means for displaying the search result together with reading information of the search result. 15. The information processing apparatus according to claim 9, further comprising a selection unit that selects a grammar according to the search result and a prediction unit that predicts a next utterance according to the grammar. 16. The information processing apparatus according to claim 9, further comprising control means for receiving character information and controlling the character information to be processed together with the recognition result of the input voice.