WO2002039425A1

WO2002039425A1 - V-mode conversation method and system using portable telephone, and v-mode center

Info

Publication number: WO2002039425A1
Application number: PCT/JP2000/007921
Authority: WO
Inventors: Tadamitsu Ryu; Kyouhei Matsuyama; Kouzou Fukushi
Original assignee: Cai Co., Ltd.
Priority date: 2000-11-10
Filing date: 2000-11-10
Publication date: 2002-05-16

Abstract

A V-mode conversation method for connection of a portable telephone to an IP server or a WWW server of a desired service type over a network through natural, adequate conversation comprising a step of storing service types and specific words sorted by type in a database (step 1), a step of receiving declaration of intention of use of the V-mode from the portable telephone of the user (step 2), a step of recognizing the inputted speech of the user and specifying the service type on the basis of the specific words sorted by type determined by word decomposition/analysis (step 3), and a step of accessing an IP server or WWW server of the specified type through a network (step 4).

Description

Specification

TECHNICAL FIELD The present invention relates to a voice-based dialogue with a mobile phone in an environment where a mobile phone can connect to a network and receive various services. On V-mode interaction methods, systems and V-mode services using a mobile phone that can connect to an IP server or www server of the desired service type through i-mode (registered trademark) and WAP services The use of mobile phones designed to perform desired input from various types of available services through natural and appropriate dialogue with mobile phones. The V-mode interaction method, system and V-mode center.

All views of art

As of the end of October 2001, the number of mobile phone subscribers had reached 4.7 million, and mobile phones have become very popular at an amazing speed. The market needs for mobile phones are also slowly changing. Due to the use of voice to talk with friends and use it as a means of business with communication tools, the number of users who are connected to a personal computer and perform overnight communication with the network is increasing. Against this background, mobile phones such as i-mode (registered trademark) and WAP services, in which mobile phones themselves have the function of connecting to a network, have emerged, and are sweeping the market.

The i-mode (registered trademark) function provided by NTT DOCOMO provides a wide variety of functions for the Internet age, such as e-mail, information content provision, and homepage browsing on a single mobile phone. It is already establishing a firm position in the industry. NTT Docomo's i-mode (registered trademark) service, which started service in February 1999, has already exceeded 200,000 service subscribers as of the end of October 1999, and is still accelerating. It is a momentum to increase it.

The functions of i-mode (registered trademark) can be broadly divided into three services. The first is the e-mail function. It is possible to send and receive e-mail to the Internet even though it is as small as 250 characters per mail, and a dedicated mail address is also provided The second is information content. In addition to news, weather forecasts, economic information such as stock prices, travel information, town and gourmet information, entertainment and other information-based services, mobile banking, ticket reservations, online trading, etc. Interactive content is also available. And, type III, you can browse the Internet homepage while displaying text-based simple display (small images can also be displayed).

At the same time as i-mode (registered trademark), the DDI Cellular Group and ID0 Telecommunications Co., Ltd. launched a cdmaOne (registered trademark) service equipped with a browser that uses a new protocol called WAP (Wireless Communication Protocol). It was realized. The content of the information service using WAP is the provision of various information contents such as news, information, amusement, ticket reservation, shopping, etc., plus e-mail (200 characters received). . From January 2000, the cdmaOne (registered trademark) service will perform packet communication at 64 kbps, which is significantly higher than the 96-kbps i-mode (registered trademark) data communication speed. As a result, the capacity of communication contents has begun to increase.

However, in any of the above, the work itself to specify the service desired by the mobile phone user is performed using a selection key having scrolling up and down (left and right) and a selection function and / or a standard numeric keypad. It was like. Therefore, when the menu of services becomes enormous, there are drawbacks that the user must scroll the screen endlessly to identify the service desired by the user, or cannot be overlooked. And even if they were found, the time it took to find the service they wanted, except for the ones they used frequently, was extremely long. Furthermore, with the miniaturization of mobile phones, the screen for displaying text information was only about 5 cm square, making it difficult or invisible for people with low vision.

There was also a method of specifying a desired service by inputting characters using the numeric keypad, but there was a drawback that if the wrong key to search for the desired service was found or it could not be remembered, it would not be possible to find it in the end. . In addition, this method of describing characters using a numeric keypad is a single sentence due to the limited number of keys. The drawback was that the key had to be pressed many times to create the character. This work was cumbersome for ordinary users and it was difficult to learn the driving method. Therefore, there has been a demand to develop a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone that enable anyone to easily find a desired service type in a short time. .

The present invention meets the above-mentioned demand, and a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone that enable anyone to easily search for a desired service type in a short time. The purpose is to provide.

The present invention further provides a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone that can reduce the number of accesses to the host computer overnight and reduce the billing fee. The purpose is to do.

The present invention further provides a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone that can reliably search for a desired service type by asking the user. Aim.

The present invention further provides a V-mode interactive method using a mobile phone capable of searching for a desired service type by querying for synonyms, synonyms, and related words when a candidate service type is not found. The purpose is to provide a V-mode conversation system using a telephone.

The present invention further aims to reduce the time by using a conventional screen display selection method when it is expected that a long time will be required for the voice reading method when there are a plurality of options. The purpose of the present invention is to provide a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone.

Disclosure of the invention

A first aspect of the present invention to achieve the above object is to connect to an IP server or a www server of a desired service type via a network through a natural and appropriate conversation with a mobile phone. A V-mode dialogue method using a mobile phone, wherein a service type that can be provided to a mobile phone user, and a type identification word used to specify the service type are stored in a database. The user's mobile phone Receiving V-mode intention indication from the user, and recognizing the voice input from the user of the mobile phone, and using the word for type identification obtained by word decomposition and word analysis to receive through the network V-mode using a mobile phone, comprising a step of specifying a service type of the service and a step of accessing an IP server or a www server of the specified service type via a network. Provide an interactive system.

As a result, the host computer obtains the content from the IP server or the www server and provides the content to the user, or causes the IP server or the www server to perform a predetermined operation for the user.

According to a first aspect of the present invention, there is provided the V-mode interactive method using the mobile phone according to the first aspect, wherein the mobile phone is provided with an external or built-in memory and / or CPU, and And a step of downloading a speech recognition software prepared on the host computer side, a service type, and a type specifying word used for specifying the service type from the host computer to the mobile phone via a wireless line. It is characterized by having.

According to a third aspect of the present invention, there is provided the V-mode interactive method using the mobile phone according to the first or second aspect, further comprising a service in which an utterance sent from a user is to be received via a network. The method is characterized in that it includes a step of asking the user if the type cannot be specified. '' According to the present invention as set forth in claim 4, in the V-mode interactive method using the mobile phone as set forth in claim 3, the process of returning to the user when no candidate is found includes synonyms, synonyms, and synonyms. It is performed by a query sentence that inquires related words.

According to a fifth aspect of the present invention, in the V-mode interactive method using the mobile phone according to the third aspect, the step of returning to the user when there are a plurality of candidates is performed on a screen display unit of the mobile phone. A feature is that a plurality of candidates are displayed, and a mobile phone user selects an optimum one from among the candidates and makes the selection by voice or key input. Thereby, additional information for specifying the service type is obtained. A second aspect of the present invention is to provide a mobile phone using a mobile phone that can connect to an IP server www server of a desired service type through a network through a natural and appropriate dialogue with the mobile phone. A mode interaction system,

Means for indicating intention to use the V-mode, conversion control means for converting voice from a user into voice signals, voice signal transmission / reception control means for transmitting / receiving voice signals to / from a host computer, and transmission from a host computer. A mobile phone comprising voice synthesis control means for generating voice by synthesizing the voice signal sent thereto;

It can be connected to many mobile phones by wireless communication, and via a network

A host computer that can be connected to an IP server or www server, and a type identification word recording unit that records type identification words used to identify service types, voice recognition software, word decomposition software, word analysis A memory having an application recording unit for recording application such as software and voice synthesis software, an audio signal transmission / reception control means for transmitting / receiving an audio signal to / from a mobile phone, and an audio signal transmitted from the mobile phone. Voice data conversion control means for converting to voice data, and calling various software from memory, voice recognition of voice data from mobile phone users, word decomposition-word analysis to extract type identification words The user of the mobile phone tries to receive through the network by means of the extraction control means of the type identification word and the extracted type identification word. Using a mobile phone comprising service type specifying control means for specifying a service type to be used, and access control means for accessing an IP server or a www server of the specified service type via a network. Provides all V-mode interactive systems.

Accordingly, the host computer obtains the content from the IP server or the www server and provides the content to the user, or causes the IP server or the www server to perform a predetermined operation for the user.

According to a third aspect of the present invention, there is provided a V-mode center of a V-mode interactive system using the above-mentioned mobile phone.

The invention will be described in more detail hereinafter with reference to preferred embodiments illustrated, which are merely examples and do not limit the scope of the invention. It should be noted that various modifications and alterations of the present invention are possible without departing from the spirit of the invention described in the appended claims.

Brief description of the meta

FIG. 1 is a flowchart of one embodiment of a V-mode interactive method using a mobile phone according to the first aspect of the present invention.

FIG. 2 is a block diagram showing an embodiment of a system for implementing the V-mode dialog system shown in FIG.

FIGS. 3 (a) and 3 (b) are block diagrams for explaining the details of the mobile phone 10 and the host computer 20 of FIG. 2, respectively.

昍昍暴 The shape of a tyrannical bear

Hereinafter, an embodiment of a V-mode interactive method using a mobile phone and a V-mode interactive system using a mobile phone according to the present invention will be described in detail with reference to the drawings. First, FIG. 1 is a flowchart of one embodiment of a V-mode conversation method using a mobile phone according to the first embodiment of the present invention, and FIG. 2 is a flowchart of the V-mode shown in FIG. FIG. 1 is a block diagram showing an embodiment of a system for implementing a virtual conversation system.

The topics dialogue method according to the first aspect of the present invention generally includes a database formation step (step 1), a V mode use intention display step (step 2), and a service type identification step (step 3). ), And an identified service execution process (step 4).

In step 1, in the V-mode interactive method using a mobile phone according to the first aspect of the present invention, first, it is assumed that the host computer that manages a system for implementing the V-mode interactive method has an interactive mode. O Build a data pace for

The database stores at least service types that can be provided to mobile phone users and type identification words used to identify service types (see Fig. 4). The service type includes, for example, news, weather forecast, stock price information, and town information in the life and information genre. Entertainment genres include fortune-telling, games, karaoke information, character distribution, and FM song information. In the guide genre, restaurants, changing train networks, and town information Information, recipes, etc. The business genres include mobile banking, mobile pile trading, airline information, life insurance information, and ticket reservations. In transaction-based service types, in general, rather than acquiring content from an IP server or WWW server and providing it to mobile phone users, transfer funds or purchase tickets at an IP server or www server For example, a service to perform predetermined work for the user will be provided.

At present, this service type is more than 300000 just registered with NTT DoCoMo, and for each type, the individual service provided by the service institution is Services are provided at the level of several times to several thousands and tens of thousands of times. In the method of the present invention, as a method for selecting a desired service from among a large number of service types, a dialog is used with a mobile phone and, therefore, with a host computer connected to the mobile phone via a wireless line. Of course, even after the desired service type has been found and connected to the IP server or WWW server, searching for individual and subdivided services within the service can also be performed by interaction with the mobile phone. It is possible.

The type identification word used to identify a service type refers to a keyword capable of distinguishing a service type from other service types as described above.

See Figure 4). For example, "news" means not only the keyword "news" but also the name of the newspaper, news agency, or television station that provided the news, or the content of news such as three-sided articles, incidents, economics, entertainment, science, sports Can also be selected. In the case of “stock price information”, not only the keyword “stock price information” but also key indicators such as the names of stock exchanges such as the First Section of the Tokyo Stock Exchange, Nasdaq, and Mothers “commonly known as“ JPY average stock price ”and“ TOPIX ”can be selected.

The key records and analyzes what words the user tried to express for a certain service type, and identified some of the most frequently used ones as “type specific "Word". For example, when a user thinks "I want to know the trend of the Nikkei Stock Average in a specific month.""EconomicIndices","EconomicNews","I want to know stock price information", "Statistics of Nikkei Stock Average", "Trend of Nikkei Stock Average in October", etc. They ask various words. In the method of the present invention, a plurality of keywords are prepared so that a desired service type can be correctly specified. In the case of general terms such as "economy" and "news", there are multiple candidates. In this case, as described later, it is possible to specify one service type by further performing an inquiry for narrowing down.

Thus, the preparation for carrying out the method of the present invention is completed. When a user uses the method of the present invention, the following approach is required.

'That is, in step 2, the user indicates the intention to use the V mode on the mobile phone. Indication of intention can be made by voice or by using input means provided on the mobile phone. For example, a key for the V mode is installed on a mobile phone having an i-mode (registered trademark) function (reference numeral 13 in FIG. 2). And it can be constructed to change the mobile phone to V mode by pressing this. That is, select the V mode in which input is performed by voice dialogue with the input means for inputting mechanically while viewing the character information displayed on the image display unit 12 used in the i-mode (registered trademark) on the mobile phone. Install a switch. Then, when the user of the mobile phone selects the V mode, the mobile phone is set to the V mode. In addition, if the mobile phone is a V-mode dedicated machine, it will automatically display the intention to use V-mode when the power is turned on.

In step 3, the host computer first recognizes the voice input from the user of the mobile phone and extracts words for type identification by word decomposition and word analysis. At this time, the type identification word used to identify the service type and the service type stored in the database of the host computer is used. The service type that is to be received via the network is specified by the extracted type specifying words.

Alternatively, these functions can be transferred from the host computer to mobile phones. By installing an external or internal memory and / or CPU, for example, it is possible to download voice recognition software, service types and type identification words prepared by the host computer from the host computer to the mobile phone via a wireless line. Can be. Then, the mobile phone itself recognizes the voice input from the user by voice, Decompose words · Extract words for type identification by word analysis. Furthermore, after identifying the service type that is to be received via the network by using the extracted type identification word, the host computer is accessed and such a service type is identified.

It can be configured to connect to an IP server or a www server. This has the effect of reducing the number of accesses from the mobile phone to the host computer, thereby reducing billing charges.

According to the method of the present invention, a speech input from a user of a mobile phone is speech-recognized. The desired service type can be specified quickly. If no candidate for the service type using such a keyword is found, candidates for synonyms, synonyms, related words, etc. are displayed on the screen display of the mobile phone, and the list of mobile phone users is displayed from the list. Can be selected by voice or key input. Alternatively, ask the user to replace it with an appropriate keyword using a query such as "The service type requested cannot be identified with XXX. Please re-express it with synonyms, synonyms, related words, etc." It can also be used.

Also, there may be a plurality of service type candidates using such a keypad. In such a case, candidates can be displayed on the screen display of the mobile phone, and the mobile phone user can select the most suitable one from the list and perform it by voice or key input. Alternatively, it is possible to prompt the user for more utterances and correctly identify the service desired by the user from other keywords. For example, if the key word “economy” is extracted, a query for narrowing down, such as “Which area of the dish do you want to know?” Is made. By replacing the underlined part with the general term that you want to narrow down and replacing it with sports, finance, entertainment, etc., you can narrow down the service types in each field.

As described above, in the conventional method such as i-mode (registered trademark), a desired service is searched for by scrolling the screen. However, this method takes too much time when the service type reaches the level of hundreds or thousands, and the user may fail. It is also conceivable to use a tree structure for service types. However, It was difficult to determine whether to enter the service category, and it had a flaw that the user could not find the target service type without considering the same way, so it could not be used in practice. According to the method of the present invention, the target service type is specified on the host computer side by interacting with the mobile phone, so that the user is extremely easy to use. In addition, there is an advantage that the number of conversations until the target service type is specified may be one to several times, and the time required is one to several seconds.

When the service type is specified in step 4, the host computer connects the mobile phone to the IP server or WWW server of the service type via the network. Thereafter, the mobile phone acquires the content from the IP server or WWW server of the service type and provides it to the user, or if it is a transaction type, the desired work for the user is i-mode ( It is the same as a mobile phone with a registered trademark function.

Next, a V-mode interactive system using a mobile phone will be described in detail with reference to FIG.

FIG. 2 is a block diagram of one embodiment of a V-mode interactive system using a mobile phone according to the second aspect of the present invention.

The V-mode interactive system 1 according to the second embodiment of the present invention generally includes a mobile phone 10 that can use the V-mode, and a host computer 20 that can be connected to the mobile phone 10 by wireless communication. It is comprised including. The host computer 20 is connected to an IP server 40 or a WWW server 42 of a desired service type via a network 30. It is connected to an IP server for mobile banking, online trading, and other trading systems via a dedicated line 32 so that no third party can intervene. Other networks can use the INN network 34 as it is. In the illustrated preferred embodiment, the system is constructed so that it can be used in conjunction with the i-mode (registered trademark) system of NTT DoCoMo. That is, the wireless communication from the mobile phone 10 is received by the relay antenna 51 managed by NTT DOCOMO, and is connected to the i-mode sensor 55 via the NTT DOCOMO packet network 53. Then, when the user of the mobile phone 10 selects the V mode with the mobile phone 10, It is connected to the V-mode center 60 having the host computer 20 and the V-mode interactive method using the mobile phone according to the first aspect of the present invention described above can be performed.o

Of course, if the V-modem 60 has its own relay antenna or satellite, it will be able to receive the radio waves of the mobile phone 10 directly. In that case, the V-mode sensor 60 can connect the mobile phone 10 with the IP server 40 and / or the WWW server 42 without going through the i-mode sensor 55. Become. '

As shown in FIGS. 2 and 3, the mobile phone 10 is equipped with a standard equipment for a mobile phone that can use the i-mode (registered trademark) function, for example, a speaker / microphone (not shown). In addition to input keys 11 such as a numeric keypad, a liquid crystal monitor 12 and the like, a button switch 13 for indicating an intention to use the V mode, a memory 14, and a CPU 16 are provided. The CPU 16 includes a conversion control unit 16a for converting a voice from a user into a voice signal, a voice signal transmission / reception control unit 16b for transmitting / receiving a voice signal to / from the host computer 20, and a host computer. The voice synthesizing control means 16 c is provided which synthesizes a voice signal transmitted from the speaker 20 and sounds it from a speaker.

As described above, the host computer 20 can be connected to many mobile phones 10 by wireless communication. In addition, it can be connected to the IP server 40 and the WWW server 42 via the network 30. The host computer 20 is generally provided with a memory 22 and a CPU 24.

The memory 22 includes a type identification word recording unit 22 a for recording a service type and a type identification word used to identify the service type, a speech recognition software and a word decomposition software-a word analysis software. It has an application recording unit 22b for recording applications such as voice synthesis software.

The CPU 24 includes an audio signal transmission / reception control means 24 a for transmitting and receiving an audio signal to and from the mobile phone 10, and an audio signal for converting an audio signal sent from the mobile phone 10 into an audio signal. Evening conversion control means 24 b and various software are called from memory 22 to recognize voice data from mobile phone users and perform word decomposition and word analysis to classify them. Service type identification that identifies the service type that the mobile phone user is trying to receive via the network, based on the type identification word extraction control means 24c that extracts the identification word, and the extracted type identification word. It is configured to include control means 24d and access control means 24e for accessing an IP server or a WWW server of the specified service type via a network.

When the user having the mobile phone 10 having the i-mode (registered trademark) function presses the button switch 13 to display the intention to use the V-mode, the audio signal from the mobile phone 10 is output in the i-mode. It is sent to V-mode Send / Receiver 60 via Send / Receiver 55. When trying to find a desired service type using the V mode of the present invention, the user can input in either a keyed or text-based dialogue. In the case of keywords, it is preferable to speak a plurality of keywords. Thereby, a desired service type can be specified accurately and in a short time.

In the V-mode center 60, the voice signal from the mobile phone 10 is received by the voice signal transmission / reception control means 24a, and further converted into voice data readable by a computer by the voice data conversion control means 24. I do. The extraction control means 24 c calls the speech recognition software from the application recording unit 22 b of the memory 22 and the type identification word from the type identification word recording unit 22 a. Then, voice data from the mobile phone user converted by the voice data conversion control means 24b is recognized. Next, the words for type identification obtained by word decomposition and word analysis are extracted, and the type of service that the user of the mobile phone 10 is going to receive via the network is identified. Subsequently, the host computer 20 of the V-mode server 60 accesses the IP server 40 or the WWW server 42 of the specified service type via the network 30.

Claims

The scope of the claims

1. A V-mode dialogue method using a mobile phone that can connect to an IP server or a www server of a desired service type via a network through a natural and appropriate dialogue with the mobile phone. hand,

Storing in a database a service type that can be provided to a mobile phone user, and a type specifying word used to specify the service type;

Receiving a V-mode indication from the user's mobile phone;

A process of recognizing a voice input from a user of a mobile phone, and specifying a service type to be received via a network by using a type specifying word obtained by word decomposition and word analysis; and

Accessing the IP server or www server of the specified service type via a network;

Mobile phone that obtains the content from the IP server or www server and provides it to the user, or allows the IP server or www server to perform predetermined work for the user. How to talk in V mode using a telephone.

2. The V-mode interactive method using a mobile phone according to claim 1, wherein the mobile phone has an external or internal memory and / or CPU, and further comprises a host. A process is included in which voice recognition software prepared on the computer side, service type, and a type identification word used to identify the service type are downloaded from the host convenience to the mobile phone via a wireless line. V-mode conversation method using a mobile phone.

3. In the V-mode interactive method using a mobile phone according to claim 1 or 2, further, the type of service to be received via the network could not be specified in the utterance sentence from the user. A V-mode dialogue method using a mobile phone, comprising a step of asking the user to listen back.

4. In the V-mode interactive method using a mobile phone according to claim 3, the process of asking the user if no candidate is found includes synonyms, synonyms, and related terms. A V-mode interactive method using a mobile phone, which is performed by a query sentence that inquires about words and the like.

5. In the V-mode interactive method using a mobile phone according to claim 3, the step of asking the user when there are a plurality of candidates includes displaying a plurality of candidates on a screen display unit of the mobile phone, The mobile phone user selects the best one from among them and makes it done by voice or key input, thereby obtaining additional information for specifying the service type. V-mode conversation method.

6. A V-mode interactive system using a mobile phone that can connect to an IP server or WWW server of a desired service type through a network through natural and appropriate dialogue with the mobile phone. ,

Means for indicating intention to use the V-mode, conversion control means for converting voice from a user into voice signals, voice signal transmission / reception control means for transmitting / receiving voice signals to / from a host computer, and host computer A mobile phone comprising voice synthesis control means for voice-synthesizing a voice signal sent from the evening and generating sound;

A host computer that can be connected to many mobile phones by wireless communication and can be connected to an IP server or www server via a network, and records the type identification words used to identify the service type Type identification word recording unit, voice recognition software ■ Word decomposition software, word analysis software, speech synthesis software, etc., a memory having an application recording unit that records application cases, and a voice signal between the mobile phone Control means for converting voice signals sent from a mobile phone into voice data, and calling various software from a memory to transmit voice from a mobile phone user. Type identification word extraction control means for recognizing data, word resolution and word analysis to extract type identification words, and extracted seeds The service type specifying control means for specifying the service type that the mobile phone user intends to receive via the network using the specification word, and the IP server or www server of the specified service type is connected to the network. Access control means for access via

To obtain the content from the IP server or www server and provide it to the user, or the IP server or www server A V-mode dialogue method using a mobile phone to perform predetermined tasks for the user.

7. The V-mode center of the V-mode interactive system according to claim 6, wherein a type-specific word recording unit for recording a type-specific word used for specifying a service type, a voice recognition software / word. A memory having an application recording unit for recording applications such as decomposition software, word analysis software, and speech synthesis software; voice signal transmission / reception control means for transmitting / receiving voice signals to / from a mobile phone; Voice data conversion control means for converting incoming voice signals into voice data, and calling various software from memory, voice recognition of voice data from mobile phone users, word decomposition and word analysis to identify types Mobile phone users will receive via the network by means of extraction control means for type-specific words for extracting word for use and the extracted type-specific words. And service type specifying control means for specifying a service type that is, and an access control means for accessing via the network to the I P server or WWW server of the identified service type,

And obtains the content from the IP server or www server and provides it to the user, or causes the IP server or www server to perform a predetermined operation for the user. Modsen evening.