Information processing device

Abstract

To provide, in voice search, a technology for performing a search using different search conditions when search results different from a user's intention are output and the user performs a search again using the same utterance.SOLUTION: An information processing device according to one aspect of the present disclosure outputs results based on a first search condition determined by speech recognition of a user's first search utterance. When, after viewing the search results, the user makes an utterance indicating that the search results differ from the user's intention, the information processing device outputs a message prompting the user to make a second search utterance. When it is determined, as a result of speech recognition of the user's second search utterance, that the second search utterance matches the first search utterance, the information processing device outputs search results again based on a second search condition different from the first search condition.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device. [Background technology]

[0002] Patent document 1 proposes a voice dialogue device that obtains search candidates based on speech recognition results and attribute information, and if the search candidates are not what the user is looking for, changes the attribute conditions of the search target according to the genre attributes selected by the user and searches again. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-97082 Summary of the Invention [Problem to be solved by the invention]

[0004] One of the objectives of the present disclosure is to provide a technology for performing a search using different search conditions when a search result different from the user's intention is output in a voice search and the user performs a search again using the same utterance. [Means for solving the problem]

[0005] An information processing device according to a first aspect of the present disclosure includes a control unit configured to: output search results based on first search criteria determined by speech recognition of a first search utterance of a user; and, if the user utters an utterance indicating that the search results differ from the user's intended search results after the output of the search results based on the first search criteria, output a message prompting the user to utter a second search utterance; and, if the second search utterance of the user is determined to match the first search utterance as a result of speech recognition of the second search utterance of the user, output another search result based on second search criteria different from the first search criteria. [Effects of the Invention]

[0006] According to the present disclosure, in a voice search, if search results different from the user's intention are output and a search is performed again using the same utterance, a search can be performed using different search conditions. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 schematically illustrates an example of a situation to which the present disclosure is applied. [Figure 2] FIG. 2 is a diagram showing an example of various types of information stored in a database. [Figure 3] FIG. 3 shows an example of a processing procedure of the information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0008] [1 Application example] FIG. 1 schematically illustrates an example of a scenario to which the present disclosure is applied. The information processing device 1 according to this embodiment includes a control unit 11. The control unit 11 outputs results based on first search criteria determined by speech recognition of a user's first search utterance. When the user, upon viewing the search results, makes a speech indicating that the search results differ from the intended search results, the control unit 11 outputs a message encouraging the user to make a second search utterance. When the second search utterance by the user is determined to match the first search utterance as a result of speech recognition, the control unit 11 is configured to output search results again based on second search criteria different from the first search criteria.

[0009] As another form of the information processing device 1 according to the above embodiment, one aspect of the present disclosure may be an information processing method that realizes all or part of the above components, a program, or a storage medium that stores such a program and is readable by a machine such as a computer. A storage medium that is readable by a machine such as a computer is a medium that stores information such as a program by electrical, magnetic, optical, mechanical, or chemical action.

[0010] (Voice Recognition) Speech recognition refers to converting speech data acquired by a speech sensor S into digital data, analyzing the digital data, and converting the digital data into text data that can be recognized by a computer. A known method, such as a technique using a hidden Markov model, may be adopted for speech recognition. In one example of this embodiment, the search utterance acquired by the speech sensor S is subjected to acoustic analysis and converted into digital data, and then phonemes are extracted using an acoustic model. Meaningful words are constructed from the extracted phonemes based on a pronunciation dictionary.

[0011] (search utterance) A search utterance is an utterance that expresses a search request from a user to the information processing device 1. In one example, the utterance is composed of a target search term and a predicate requesting a search, such as "Search for XX" or "Display XX." In another example, the search utterance may simply be an utterance of the target search term alone. The control unit 11 determines first search conditions by performing voice recognition on the first search utterance uttered initially.

[0012] (Search conditions) The search conditions are conditions specified when requesting search results from the search server SS, and are determined by speech recognition of the search utterance. The method of determining the search conditions by speech recognition of the search utterance is not particularly limited and may be selected appropriately depending on the embodiment. The search conditions may include specification of at least one of a search term and a search category.

[0013] (A) Search terms Search terms are keywords that serve as queries for searches and are determined by speech recognition of the search utterance. When determining search terms, words corresponding to the speech of the search terms are identified from the search utterance, and words that at least partially contain those words are extracted. The extracted words may include multiple homonyms, and the method for determining which words to use as search terms may be selected appropriately depending on the embodiment. In one example, when there are multiple homonyms, priorities may be set for each word, and search terms may be determined based on the priorities. In another example, the actual search frequency for each homonym may be tallied, and search terms may be determined based on the frequencies. When determining search terms based on frequency, for example, when someone utters "Search for Tarumizukou," speech recognition is used to first extract "Tarumizukou" as a word corresponding to the speech of the search term. Homonyms corresponding to the speech of the search term include "Tarumi Port" and "Tarumi School," but "Tarumi Port" is determined as the search term because it was searched more frequently.

[0014] (B) Search Category A search category refers to an item used to classify information to be searched. There may be multiple categories, and the method of selecting which category to use as the search category may be selected appropriately depending on the embodiment. In one example, when there are multiple categories, a priority may be set for each category, and the search category may be determined according to the priority. In another example, the frequency of actual searches for each category may be tallied, and the search category may be determined according to the frequency. For example, When the search target includes place names and facilities, possible categories include the name, facility genre, area, etc. In this case, when determining the search category based on priority, the facility genre category has a high priority, so this category is specified as the search category.

[0015] If multiple words that at least partially contain a word corresponding to the speech of the search term are extracted, multiple items may exist in the determined search category, and which item to select may be selected as appropriate depending on the embodiment. For example, consider the case where the search target includes place names and facilities, and the facility genre is specified as the search category, and the user utters the command "Search for Tarumizukou." First, "Tarumizukou" is extracted as a word corresponding to the speech of the search term using speech recognition. Words corresponding to the speech of the search term include homonyms such as "Tarumi Port" and "Tarumi School," and items in the facility genre include "port" and "school facilities." The control unit 11 incorporates "school facilities" within the facility genre into the search criteria using a predetermined determination method (e.g., priority).

[0016] The information processing device 1 communicates with the search server SS via the communication interface 13 and transmits a search request based on the determined first search criteria. Upon receiving the search request, the search server SS accesses the database DB in which the information to be searched is stored, specifies the search criteria, and acquires the information to be searched.

[0017] FIG. 2 shows a schematic example of various types of information stored in the database DB. The information to be searched may be stored in any database format. When the information to be searched is stored in a relational database format, the columns of the table stored in the database DB may indicate search categories. In the example of FIG. 2, the search targets include place names and facilities, and also include names (corresponding to search terms), genres, regions, etc. The search server SS acquires information from the database DB by performing a search by specifying at least one of search terms and search categories based on the search conditions included in the search request received from the information processing device 1. The search server SS transmits the acquired information to the information processing device 1.

[0018] The information processing device 1 presents the search results acquired from the search server SS to the user by displaying them on the output device 15. When the information processing device 1 determines that the user who has checked the search results has made an utterance indicating that the search results are different from the intended search results, it outputs a message prompting the user to make a second search utterance. The utterance indicating that the search results are different from the intended type of search results may be, for example, an utterance having a negative meaning, such as "different" or "wrong."

[0019] Furthermore, the content of the message encouraging the user to make a second search utterance is not particularly limited as long as it is a message that encourages the user to speak again, and may be selected appropriately depending on the embodiment. For example, it may be a phrase such as "Please speak again." The method of outputting the message is also not particularly limited, and may be selected appropriately depending on the embodiment. For example, it may be displayed on a display or output as a voice message.

[0020] (Search utterance match determination) A method for determining whether a first search utterance and a second search utterance match may be selected as appropriate depending on the embodiment. In one example, if the phonemes of the search utterance extracted by speech recognition are identical between the first search utterance and the second search utterance, it may be determined that the search utterances match. Alternatively, if the first search utterance and the second search utterance at least partially match, it may be determined that the search utterances match. In one example, if, as a result of speech recognition of the search utterances, the words corresponding to the extracted search terms are identical between the first search utterance and the second search utterance, it may be determined that the search utterances match.

[0021] A method for determining a second search condition when it is determined that a first search utterance and a second search utterance match The method is not particularly limited and may be selected appropriately depending on the embodiment. In one example, the second search criteria may include a search term that is different from the search term included in the first search criteria. In another example, if the search criteria include a search category, the second search criteria may include a search category that is different from the search category included in the first search criteria. Furthermore, while specifying the same search category, an item that is different from the item in the search category included in the first search criteria may be specified and included in the second search criteria.

[0022] [2 Configuration Examples] 1, the information processing device 1 according to this embodiment is a computer in which a control unit 11, a storage unit 12, a communication interface 13, an external interface 14, and an output device 15 are electrically connected. The control unit 11 includes a CPU (Central Processing Unit), a RAM, The storage unit 12 includes a random access memory (Random Access Memory), a read-only memory (ROM), etc., and is configured to execute any information processing. The storage unit 12 may be configured, for example, by a hard disk drive, a solid-state drive, etc. In the present embodiment, the storage unit 12 stores a program 81. The program 81 is a program for causing the information processing device 1 to execute information processing related to a search based on a search utterance obtained by voice recognition. The program 81 includes a series of instructions for the information processing. The communication interface 13 is configured to perform wired or wireless data communication via a network. The communication interface 13 may be configured, for example, by a wired local area network (LAN) module, a wireless LAN module, etc. In the present embodiment, the information processing device 1 may use the communication interface 13 to perform data communication with another computer (e.g., a search server SS) via a network. The external interface 14 may be, for example, a universal serial bus (USB) port, a dedicated port, a wireless communication port, etc., and is configured to connect to an external device via a wired or wireless connection. In the present embodiment, the information processing device 1 may be connected to a sound sensor S for acquiring sound via the external interface 14. The audio sensor S may be provided as an input device within the information processing device 1. The audio sensor S may be configured, for example, by a microphone. The output device 15 is a device for outputting, for example, a display, a speaker, etc. In this embodiment, search results, messages, etc. are output to the display.

[0023] Regarding the specific hardware configuration of the information processing device 1, components can be omitted, replaced, or added as appropriate depending on the embodiment. For example, the control unit 11 may include multiple hardware processors. The hardware processors may be a microprocessor, a field-programmable gate array (FPGA), a digital signal processor (DSP), The information processing device 1 may be configured with a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), etc. The information processing device 1 is designed specifically for the services provided. In addition to a computer, a general-purpose server device or the like may also be used.

[0024] [3 Example of operation] 3 shows an example of a processing procedure of the information processing device 1 according to this embodiment. The following processing procedure is an example of an information processing method executed by a computer.

[0025] (Steps S101 to S105) In step S101, the control unit 11 acquires a voice input made by the user. In step S102, the acquired voice input is recognized as a first search utterance, and in step S103, a first search condition is determined. In step S104, the control unit 11 transmits a search request to the search server SS based on the determined first search condition. In step S105, the control unit 11 displays the search results transmitted from the search server SS on the output device 15.

[0026] (Steps S106 to S107) In step S106, the control unit 11 determines whether or not the user, who has confirmed the displayed search results, has made an utterance indicating that the search results are different from the intended search results. If the utterance has not been made, the processing procedure of the information processing device 1 according to this operation example is terminated. If the utterance has been made, the process proceeds to step S107, where the control unit 11 outputs a message to the user prompting them to make a second search utterance.

[0027] (Steps S108 to S113) In step S108, the control unit 11 again acquires a voice input made by the user. In step S109, the acquired voice input is recognized as a second search utterance. In step S110, the control unit 11 determines whether the first search utterance and the second search utterance match. If it is determined that they match, the control unit 11 proceeds to step S111 and determines second search conditions that are different from the first search conditions. In step S112, the control unit 11 again sends a search request to the search server SS based on the determined second search conditions. In step S113, the control unit 11 again displays the search results sent from the search server SS on the output device 15, and ends the processing procedure of the information processing device 1 according to this operation example.

[0028] On the other hand, if it is determined in step S110 that the first search utterance and the second search utterance do not match, the control unit 11 returns the process to step S103. In step S103, the control unit 11 treats the second search utterance as a new first search utterance and determines first search conditions based on the search terms included in the search utterance. Then, the control unit 11 executes the search process from step S104 onwards again.

[0029] The control unit 11 may execute the processing procedures of steps S101 to S113 each time a search utterance is made by the user.Furthermore, after a search is performed, the control unit 11 may execute the processing procedures of step S107 and thereafter each time a user who has checked the search results makes an utterance indicating that the search results are different from the intended search results.

[0030] [Features] In conventional searches using voice recognition, if a search utterance made again due to a recognition error is identical or similar to the previous utterance, the same voice recognition is performed as the previous utterance, resulting in search results that differ from the user's intention. In contrast, in this embodiment, after the user makes an utterance indicating that the search results are different from the intended search results, it is determined whether the content of the reuttered utterance is identical to the previous utterance. If it is determined that they are identical, a search is performed by specifying search conditions different from the previous utterance. This makes it possible to avoid outputting the same search results when the same utterance is made, and to quickly present search results of the type intended by the user. [4 Variations] Although the embodiments of the present disclosure have been described in detail above, the above description is merely an example of the present disclosure in all respects. It goes without saying that various improvements or modifications can be made without departing from the scope of the present disclosure. The processes and means described in the present disclosure can be freely combined and implemented as long as no technical contradiction occurs. [Explanation of symbols]

[0031] 1. Information processing device, 11. Control unit, 12. Storage unit, 13. Communication interface, 14. External interface, 15··Output device, 81··Program, S··Speech sensor, SS··Search server, DB··Database

Claims

[Claim 1] An information processing device including a control unit, The control unit outputting search results according to the first search criteria determined by speech recognition of the user's first search utterance; and after outputting the search results based on the first search criteria, if the user utters an utterance indicating that the search results differ from the search results intended by the user, output a message encouraging the user to utter a second search utterance, and when it is determined that the second search utterance of the user matches the first search utterance as a result of speech recognition of the second search utterance of the user, output another search result based on second search criteria different from the first search criteria; configured to perform Information processing device.

Images