JP2023176156A - Information processing device - Google Patents

Abstract

To provide an information processing device that can improve convenience of speech recognition.SOLUTION: An information processing device 100 which consists of an information server connected via a network to an electronic device that allows a user to input speech, includes speech input means 110, schedule acquisition means 120, intention estimation means 130, and output means 140. The speech input means 110 acquires an utterance content of the user. The schedule acquisition means 120 acquires action schedule information of the user registered in a schedule. When searching for facility information based on the utterance content, the intention estimation means 130 refers to facility information related to the action schedule information registered in the schedule within a period from present to a predetermined time ahead, and performs intention understanding of the utterance content. The output means 140 outputs results of the intention understanding.SELECTED DRAWING: Figure 2

Description

The present invention relates to an information processing device.

When you talk to an agent (voice dialogue service), it responds with voice and searches for destinations and information. Since all operations including setting the destination can be performed by voice, it is comfortable and convenient while driving a car where manual input using a keyboard or the like cannot be performed.

Various techniques have been developed to improve the accuracy of voice recognition by agents. For example, Patent Document 1 discloses a speech recognition system and the like that can improve the accuracy of speech recognition even with a small number of speech inputs from the user by utilizing past utterance history data for intention understanding.

Japanese Patent Application Publication No. 2016-24652

As in the technology disclosed in Patent Document 1, past utterance history data is actively used to understand intentions, while future data is hardly used to understand intentions. For example, regarding a facility search, if a plurality of facility search results based on voice recognition match, an additional question to the user is required to further specify the facility. This requires multiple utterances, which may be troublesome for the user.

The present invention has been made in view of such problems, and an object of the present invention is to provide an information processing device that can improve the convenience of voice recognition.

An information processing apparatus according to one aspect of the present invention includes a voice input means, a schedule acquisition means, an intention estimation means, and an output means. The voice input means acquires the content of the user's utterance. The schedule acquisition means acquires the user's action schedule information registered in the schedule. When searching for facility information based on the content of the utterance, the intention estimation means refers to the facility information related to the action plan information registered in the schedule for a period from now to a predetermined time period, and understands the intention of the content of the utterance. conduct. The output means outputs the result of intention understanding.

According to this aspect, since the user's intention is understood to specifically specify facility information by taking into account the user's future action schedule, the convenience of voice recognition is improved.

In the above aspect, when a plurality of pieces of action plan information are listed as candidates for intention understanding, it is preferable that action plan information for the nearer future is configured to be given priority for intention understanding.

According to this aspect, among a plurality of pieces of action plan information, facility information related to action plan information that is likely to be the user's intention is displayed at the top, thereby further improving the accuracy of intention understanding by voice recognition. be able to.

According to the present invention, it is possible to provide an information processing device that can improve the convenience of voice recognition.

FIG. 1 is a diagram showing an example of a schematic configuration of an information processing apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram showing an example of the functional configuration of an information processing device according to an embodiment of the present invention. FIG. 3 is a diagram illustrating an example of speech recognition using schedule information by an information processing apparatus according to an embodiment of the present invention.

Preferred embodiments of the present invention will be described with reference to the accompanying drawings. In addition, in each figure, those with the same reference numerals have the same or similar configurations. Hereinafter, the present invention will be explained in detail with reference to the drawings.

FIG. 1 is a diagram showing an example of a schematic configuration of an information processing apparatus according to an embodiment of the present invention. The information processing device 100 has the function of an agent (voice dialogue service) that responds with voice when spoken to and searches for destinations and information. The information processing apparatus 100 includes, for example, an electronic device 1 to which a user can input voice input, an information server 2 connected via a network N, and the like. The electronic device 1 may be an in-vehicle device such as a head unit that can be mounted on a vehicle, or may be a mobile terminal such as a smartphone.

The electronic device 1 has a control section 11, a storage section 12, etc. as a functional configuration. Similarly, the information server 2 has a control section 21, a storage section 22, etc. as a functional configuration. Note that the components of the electronic device 1 and the information server 2 are not limited to the illustrated example, and arbitrary components may be added as necessary.

The electronic device 1 has a function as a car navigation device. The control unit 11 has a function of executing various processes in the electronic device 1, such as a voice input means 110, which will be described later, in addition to navigation processing. The storage unit 12 stores various data processed by the control unit 11.

The control unit 21 has a function of executing various processes in the information server 2. For example, the control unit 21 has a function of receiving data transmitted from a vehicle and executing various processes in the electronic device 1. The storage unit 22 stores various data. Control units 11 and 21 and storage units 12 and 22 of electronic device 1 and information server 2 constitute voice input means 110, schedule acquisition means 120, intention estimation means 130, output means 140, etc., which will be described later.

FIG. 2 is a block diagram showing an example of the functional configuration of the information processing device 100 according to the embodiment of the present invention. As shown in FIG. 2, the information processing device 100 includes a voice input means 110, a schedule acquisition means 120, an intention estimation means 130, and an output means 140.

The voice input means 110 acquires the content of the user's utterance by converting the user's voice input from a microphone or the like of the electronic device 1 into an electrical voice waveform signal. The schedule acquisition unit 120 refers to a schedule DB in which a user's schedule can be registered, and acquires the user's action schedule information registered in the schedule. The schedule DB may be included in the storage unit 22 of the information server 2 that constitutes the information processing device 100, or may be included in a server different from the information server 2.

The intention estimation means 130 understands the intention of the utterance content acquired by the voice input means 110. For example, when searching for facility information based on the content of utterances, the intention estimation means 130 may search facilities related to action plan information registered in a schedule from now to a predetermined time (for example, within one week). Refer to the information and understand the intent of the utterance. The intention estimating means 130 is configured to allow the user to arbitrarily set the range of schedules to be referenced. For example, the user can set a predetermined time period through the voice input means 110 to refer to the schedule within a period from the present time.

The output means 140 uses the electronic device 1 to output the result of intention understanding. For example, the result of intention understanding can be displayed on the monitor of electronic device 1, or the result of intention understanding can be uttered from the speaker of electronic device 1.

FIG. 3 is a diagram illustrating an example of speech recognition using schedule information by the information processing apparatus 100 according to an embodiment of the present invention. As shown in FIG. 3, one of the characteristics of the information processing apparatus 100 functioning as an agent is that when performing a facility search using voice recognition, data registered in the schedule is referenced and utilized for understanding the intention.

For example, there is an izakaya chain that operates multiple stores (store B, store C, store D, etc.) using the store name "A Izakaya," and the schedule includes activity schedule information such as "January 1st 18:00. In the case where "A Izakaya B store" is registered, when the user utters "A Izakaya", "Izakaya", or "Drinking party", "A Izakaya B store" is displayed at the top.

The proper noun "A Izakaya" is a keyword that partially matches "A Izakaya B store" registered in the schedule. The common noun "Izakaya" is a keyword that partially matches "A Izakaya B store" registered in the schedule.

“Izakaya” and “drinking party” are examples of keywords related to “A Izakaya B store”. "Izakaya" is included in subdivision number 7651 "Bars and beer halls" of the Japan Standard Industrial Classification, which is the industry to which the facility of "A Izakaya B store" belongs, and is a keyword closely related to "A Izakaya B store". . “Drinking party” is the main purpose of use of the facility of “A Izakaya B store” registered in the schedule, and is a keyword closely related to “A Izakaya B store”.

According to the information processing apparatus 100 of the present embodiment configured as described above, the user's future action schedule is taken into account to understand the user's intention to specifically specify the facility information, thereby improving the convenience of voice recognition. will improve. The information processing device 100 that can improve the convenience of voice recognition is comfortable and convenient while driving a car where manual input using a keyboard or the like cannot be performed.

When a plurality of pieces of action plan information are selected as candidates for intention understanding, it is preferable that action plan information in the near future be prioritized for intention understanding. For example, if "A Izakaya B store at 18:00 on January 1st" and "A Izakaya C store at 17:00 on January 2nd" are both registered in the schedule within the period from now to a predetermined time, When the user utters "A-izakaya," "izakaya," and "drinking party," action schedule information for January 1st, action schedule information for January 2nd, and a plurality of action schedule information are listed as candidates for understanding the intention.

In such a case, the system is configured to give priority to the action schedule information for January 1st, which is in the nearer future, to understand the intention and display "A Izakaya B store" higher than "A Izakaya C store". Preferably. According to this aspect, among a plurality of pieces of action plan information, facility information related to action plan information that is likely to be the user's intention is displayed at the top, thereby further improving the accuracy of intention understanding by voice recognition. be able to.

The information processing device 100 is configured to be able to select behavior schedule information to be used for intention understanding and behavior schedule information not to be used for intention understanding, by referring to attributes given to behavior schedule information registered in the schedule. may have been done. For example, if action schedule information is divided into an attribute of "public information" that can be referenced by other users who use the schedule database, and an attribute of "private" that can be referenced only by the user, "private" It may be configured such that only the attributes of are used for understanding the intention.

If the attribute of "work" or "off" is assigned to the action schedule information by tag classification by the user, etc., the configuration may be such that only the "off" attribute is used to understand the intention. As shown in FIG. 2, the information processing device 100 may further include determining means 150 for determining the attributes of proper nouns.

The embodiments described above are intended to facilitate understanding of the present invention, and are not intended to be interpreted as limiting the present invention. Each element included in the embodiment, as well as its arrangement, material, conditions, shape, size, etc., are not limited to those illustrated, and can be changed as appropriate. Further, it is possible to partially replace or combine the structures shown in different embodiments.

DESCRIPTION OF SYMBOLS 1... Electronic device, 2... Information server, 11, 21... Control unit, 12, 22... Storage unit, 100... Information processing device, 110... Voice input means, 120... Schedule acquisition means, 130... Intention estimation means, 140... Output means, 150...determination means.

Claims (1)

a voice input means for acquiring the content of the user's utterance;
a schedule acquisition means for acquiring action schedule information of a user registered in the schedule;
When searching for facility information based on the content of the utterance, the intent of the content of the utterance is understood by referring to facility information related to the action plan information registered in the schedule for a period from now to a predetermined time ahead. Intention estimation means;
An information processing device comprising: output means for outputting the result of the intention understanding.