JP2003319070A - Voice response method and unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a highly convenient voice response method and unit in which a contents provider can register a key word freely. <P>SOLUTION: In the voice response unit for providing contents to the user of a voice communication terminal 1 connected through a voice communication network 6 by voice interaction, an objective keyword of voice recognition received from the terminal 2 of the contents provider connected through a network 4, and reference information for accessing contents corresponding to the keyword are registered on a table 10. When a keyword uttered by the user has been registered on the table 10, a URL corresponding to the keyword is acquired and then the contents concerning to that URL are acquired through the network 4 and provided to the user. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technology for providing a user with contents by a voice dialogue with a voice communication terminal connected via a voice communication network, and particularly to the Internet (IP).
Network) to provide a voice portal service using a network.

[0002]

2. Description of the Related Art In conventional voice portal services, services selectable by voice keyword are contents (weather information, news information, stock price information, road traffic information, fortune telling) registered in advance by service providers (communication carriers, etc.). , Transfer information, gourmet information, etc.). When accessing content created by a content provider other than the service provider, dial a numeric code or assign multiple characters (A, B, C, etc. to button 1) to one button on the telephone. URL (Uniform Resource Lo)
The operability was poor because it was specified.

[0003]

In the conventional method, the menu that the voice portal service user can easily select by uttering an easy-to-understand keyword is limited to the menu registered in advance by the service provider. There has been no means for easily accessing the content created by the content provision such as WWW (World Wide Web). Therefore, there has been a demand for a mechanism that allows content providers to freely register keywords. In addition, when such a mechanism is realized, a problem is to prevent a situation where the keyword is well-worded and is not registered for purposes other than the original purpose.

The present invention has been made in view of the above circumstances, and a first object thereof is to provide a highly convenient voice response method and apparatus in which a content provider can freely register a keyword. To do.

A second object of the present invention is to provide a voice response method and device which can be smoothly operated by automatically eliminating unnecessary registration keywords.

[0006]

In order to achieve the above object, in a voice response device which provides contents to a user by voice dialogue with a voice communication terminal connected via a voice communication network, the voice response device is provided via the network. Registered in the storage means of the voice response device, the keyword which is the target of voice recognition and the reference information for accessing the content corresponding to the keyword via the network, which is received from the terminal of the content provider connected by When providing the content to the user, the user's utterance is voice-recognized, and if the recognized keyword is registered in the storage means, the reference information corresponding to the keyword is acquired, and the reference information is used as the reference information. Based on that, the content corresponding to the keyword is acquired through the network and the content is provided to the user. Suggest things.

According to the present invention, the keyword and the reference information to the content corresponding to the keyword can be freely registered from the terminal of the content provider, so that the content provider can create the content and perform the registration work. In response to the user's utterance of the keyword, the content corresponding to the keyword can be provided. In this way, the content provider can freely register the content, not the service provider, which is highly convenient.

Further, it is proposed to delete the registration of the content registered in the storage unit when the user does not access the content for a predetermined period of time.

According to the present invention, registration information relating to contents that are not accessed by a user for a predetermined period is automatically deleted. Therefore, registration other than the original purpose, such as keyword stock, can be eliminated and smooth operation can be performed. it can. Further, since only the necessary registration information is stored in the storage means, it is not necessary to have a large capacity as the storage means, and an improvement in the access speed when accessing the registration information from the storage means is expected. it can.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION A voice response service system in a voice portal site according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of the service system, and FIG. 2 is a detailed configuration diagram of the service system.

As shown in FIG. 1, this service system includes a voice communication terminal 1 such as a fixed telephone or a mobile telephone housed in a voice communication network 6 and a personal computer (P).
C) etc., the information terminal 2 of the content provider, and a voice response device (IVR: Interactive Voice).
The e-reception system 3 includes a network 4 such as the Internet (IP network), and a content server 5 in the network 4. A characteristic point of the present invention is that a content provider uses a content provider terminal 2 and a URL (Uniform Resource) of content scenarios via the network 4.
(e Locator) and a simple keyword for specifying the URL by voice recognition are freely registered, which enables the user to specify the content by uttering the simple keyword.

As shown in FIG. 2, the voice response device 3 includes an interface processing unit 13 that receives a connection request for the voice communication terminal 1, a communication control unit 14 that performs incoming call processing and call path establishment, and a designated guidance file. A voice reproducing unit 18 for reproducing, a basic menu scenario file 19, a voice recognizing unit 16 for monitoring user's utterance and receiving voice and recognizing words, and a basic menu scenario file 19 are acquired to transmit main menu guidance. The scenario execution environment unit 1 that gives an instruction to obtain the content scenario of the specified URL and performs a process of requesting the update of the last access date.
5 and an IP communication controller 17 that acquires designated content scenarios from the content server 5. Further, the voice response device 3 registers the URL corresponding to the keyword in the personal content management table 10, searches the personal content management table 10 by the keyword, extracts the registered URL, and has a content processing function of updating the last access date. The management unit 11 and the access management unit 12 that manages / deletes the last access date are provided.

The voice response device 3 uses a content scenario, a voice file of voice guidance to be sent to the user, and a user program as necessary to provide the content. The content scenario describes a voice file name and a procedure for providing services such as processing to be performed for each user operation (PB operation, etc.). The content means a service provided to the user by executing this content scenario in the voice response device 3.

Next, a procedure for the content provider to register a keyword and a URL corresponding to the keyword in this system will be described with reference to the flowchart of FIG. It is assumed that the content provider has already registered the content scenario in the content server 5.

The content provider activates a browser on the information terminal 2 (step S1). Next, the content provider requests the screen display for registering the URL and the keyword from the browser to the voice response device 3 (step S2).

In the voice response device 3, the IP system communication control unit 17 uses the content provider's HTTP (Hypertext).
The request by the Transfer Protocol is received (step S3). Next, the content management unit 11 of the voice response device 3 returns the screen for registering the URL and the keyword to the information terminal 2 (step S4).

Next, the content provider confirms the screen display for registering the URL and the keyword displayed on the information terminal 2 (step S5), sets the keyword of the content scenario and the URL corresponding to the keyword, and The information is returned to the voice response device 3 (step S6).

The content management unit 11 of the voice response device 3
Receives the URL corresponding to the keyword from the information terminal 2 (step S7). Next, the content management unit 11
Confirms whether the received designated keyword has already been registered in the personal content management table 10 (step S8). If the same keyword is already registered in the personal content management table 10, the fact that registration is not possible is returned to the information terminal 2 (step S9). On the other hand, if the same keyword is not registered in the personal content management table 10, the content management unit 11 registers the keyword and the URL corresponding to the keyword in the personal content management table 10 (step S10). Then, the content management section 11 returns the content registration completion display to the information terminal 2 of the content provider (step S1).
1).

The information terminal 2 of the content provider receives the completion of content registration from the voice response device 3 and displays it (step S12).

Through the above steps, the keyword and the URL corresponding to the keyword are registered in the voice response device 3 using the information terminal 2 of the content provider.

Next, the procedure in which the user designates the content by using the keyword registered by the above procedure and enjoys the service from the voice response device 3 will be described with reference to FIGS. 4 to 6.

First, when the user dials the connection number using the voice communication terminal 1 (step S21), the interface processing unit 13 of the voice response device 3 receives the connection request (step S22), and the communication control unit 14 receives the call. Processing and call path establishment are performed (step S23). In response to this, the voice communication terminal 1 also establishes a communication path (step S24).

Next, the scenario execution environment unit 15 acquires the basic menu scenario file 19 (content scenario including an explanation of the operating method) (step S25).
Then, the scenario execution environment unit 15 instructs the audio reproduction unit 18 to send out the main menu guidance according to the basic menu scenario file 19 (step S26).
An example of this main menu guidance is "○○○
If you want to hear the content registered by (content provider name), please say XXX, and if you want to △△△, please say ☆☆☆. Is mentioned. In response to the instruction, the voice reproducing unit 18 reproduces the designated guidance (step S27). The user receives the main menu guidance from the voice response device 3 (step S28).

Next, the scenario execution environment unit 15 of the voice response device 3 instructs the voice recognition unit 16 to receive the voice uttered by the user according to the basic menu scenario file 19 (step S29), and the voice recognition unit 16 The utterance of the user is monitored (step S30).

When the user selects a content scenario registered by the content provider by speaking (step S3)
1), the voice recognition unit 16 receives the voice and recognizes a word (step S32).

Next, the scenario execution environment unit 15 instructs the voice reproducing unit 18 to output the guidance corresponding to the uttered item according to the basic menu scenario file 19 (step S33), and the voice reproducing unit 19 The designated guidance file is reproduced (step S34). Here, as an example of the guidance, "Please say the keyword that specifies the content you want to listen to."
That can be mentioned. The user receives the guidance for prompting the keyword utterance from the voice response device 3 (step S35).

Next, the scenario execution environment section 15 follows the basic menu scenario file 19 and recognizes the voice recognition section 16.
The user is instructed to receive the voice (step S3
6) The voice recognition unit 16 monitors the user's utterance (step S37).

When the user utters the keyword of the designated content (step S38), the voice recognition unit 16 receives the voice and recognizes the word (step S39).

Next, the scenario execution environment unit 15 requests the content management unit 11 to specify the URL corresponding to the keyword according to the basic menu scenario file 19 (step S40). The content management unit 11 searches the personal content management table 10 using the keyword and extracts the registered URL corresponding to the keyword (step S41). The content management unit 11 confirms whether the keyword is registered (step S4).
2) If the content is not registered, reply to the execution environment unit 15 that there is no target content (step S4).
3) The scenario execution environment unit 15 gives an instruction to send error guidance (step S44).

On the other hand, when the keyword is registered, the scenario execution environment section 15 instructs the IP communication control section 17 to acquire the content scenario of the designated URL (step S45). IP communication control unit 17
Acquires a content scenario from the designated URL (step S46). In addition, the scenario execution environment unit 15
Requests the content management unit 11 to update the last access date of the keyword (step S4).
7). The content management unit 11 updates the last access date according to the request (step S48). Then
The scenario execution environment unit 15 gives guidance to the user according to the acquired content scenario (step S
49).

After that, the voice response device 3 performs the process based on the content scenario (step S50), and the user enjoys the service based on the content scenario (step S51).

Next, the procedure for automatically deleting the keyword registered from the information terminal 2 of the content provider will be described with reference to the flowchart of FIG. In this automatic deletion procedure, the access management unit 12 of the voice response device 3 regularly monitors the last access date of each keyword in the personal content management table 10, and automatically deletes the record for the keyword that has not been accessed for a certain period. To do.
Hereinafter, the procedure will be described in detail.

First, the access management unit 1 of the voice response device 3
2 is automatically activated at regular intervals (step S61), the access management unit 12 causes the personal content management table 10
The last access date of the record (step S6)
2). Then, it is determined whether the difference between the current day (current date) and the last access date (the number of days of no access) is greater than or equal to a predetermined threshold value (step S63). Here, the threshold value is a value that the voice portal service provider sets in the voice response device 3 by operation.

If the difference between the current day (current date) and the last access date is equal to or more than the predetermined threshold value, the corresponding record in the personal content management table 10 is deleted (step S6).
4).

Then, it is determined whether or not the corresponding record is the last record (step S65), and if it is the last record, the process ends (step S66). If it is not the last record, the next record is acquired and the threshold value is checked. Therefore, the process returns to step S62 and steps S62 to S66 are repeated.

As described above, in this embodiment, since the keyword and the URL of the content corresponding to the keyword can be freely registered from the information terminal 2 of the content provider, the content provider can create the content and perform the registration work. By performing the above, the content corresponding to the keyword can be provided in response to the user's utterance of the keyword. In this way, the content provider can freely register the content, not the service provider, which is highly convenient.

Further, in the present embodiment, the registration information relating to the content that is not accessed by the user for a predetermined period is automatically deleted, so registration other than the original purpose, such as keyword stock, is excluded, and smooth operation is performed. be able to. Further, since only the necessary registration information is stored in the personal content management table, it is not necessary to use a large-capacity table as the table, and it is possible to expect an improvement in the access speed when accessing the registration information from the table. .

Although the embodiment of the present invention has been described above, the present invention is not limited to this. The scope of the invention is indicated by the claims, and various modifications are included in the invention.

For example, in the above-mentioned embodiment, the Internet is exemplified as the network 4 for connecting the voice response device 3 and the information terminal 2 of the content provider, but other networks may be used.

[0040]

As described in detail above, according to the present invention,
Since the keyword and the reference information to the content corresponding to the keyword can be freely registered from the terminal of the content provider, the content provider creates the content and performs the registration work, so that the user can utter the keyword. On the other hand, it is possible to provide the content corresponding to the keyword. In this way, the content provider can freely register the content, not the service provider, which is highly convenient.

Further, since the registration information relating to the content that is not accessed by the user for a predetermined period is automatically deleted, registration other than the original purpose such as keyword stock can be eliminated and smooth operation can be performed. Further, since only the necessary registration information is stored in the storage means, it is not necessary to have a large capacity as the storage means, and an improvement in the access speed when accessing the registration information from the storage means is expected. it can.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a service system.

[Fig. 2] Detailed configuration diagram of the service system

FIG. 3 is a flowchart illustrating a procedure of keyword registration.

FIG. 4 is a flowchart illustrating a service providing procedure.

FIG. 5 is a flowchart illustrating a service providing procedure.

FIG. 6 is a flowchart illustrating a service providing procedure.

FIG. 7 is a flowchart illustrating a procedure for automatically deleting registered keywords.

[Explanation of symbols]

1 ... Voice communication terminal, 2 ... Information terminal, 3 ... Voice response device,
4 ... Network, 5 ... Content server, 6 ... Voice communication network, 10 ... Personal content management table, 11 ... Content management section, 12 ... Access management section, 13 ... Interface processing section, 14 ... Communication control section, 15 ... Scenario execution Environment part, 16 ... Voice recognition part, 17 ... IP communication control part, 18 ... Voice reproduction part, 19 ... Basic menu scenario files,

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G10L 15/10 G10L 3/00 551G 15/22 571T H04M 3/493 531W 3/50 R F term (reference) 5D015 AA04 KK04 5D045 AB26 5K015 AA06 AA07 AB01 AB02 AD01 AD02 5K024 AA71 AA76 BB01 BB02 CC11 DD01 EE09 FF04

Claims (4)

[Claims] 1. A voice response method in a voice response apparatus for providing content to a user by voice dialogue to a voice communication terminal connected via a voice communication network, the content provider connected via a network. Register the keyword that is the target of voice recognition and the reference information for accessing the content corresponding to the keyword via the network, which is received from the terminal, in the storage means of the voice response device, and provide the content to the user. In doing so, voice recognition of the user's utterance is performed, and if the recognized keyword is registered in the storage means, reference information corresponding to the keyword is acquired, and content corresponding to the keyword is acquired based on the reference information. Is obtained via a network, and the content is provided to the user. A voice response method. 2. The voice response method according to claim 1, wherein the registration related to the content is deleted when the user does not access the content registered in the storage unit for a predetermined period. 3. A voice response device for providing content to a user by voice dialogue with a voice communication terminal connected via a voice communication network, the voice response device receiving from a terminal of a content provider connected via a network. A storage unit that stores a keyword to be voice-recognized and reference information for accessing content corresponding to the keyword via a network, a voice recognition unit that voice-recognizes a user's utterance, and the recognized keyword And a reference information acquisition means for acquiring reference information corresponding to the keyword when the content is registered in the storage means, and a content corresponding to the keyword is acquired through the network based on the acquired reference information and the content is used. A voice response device comprising: a content providing unit for providing a person. 4. The voice response according to claim 3, further comprising an access management unit that deletes the registration of the content registered in the storage unit when the user does not access the content for a predetermined period. apparatus.