JP3587976B2 - Information output apparatus and method, and recording medium recording information output program - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information output apparatus and method for outputting information to a user by voice, and more particularly, to an information output apparatus and method for outputting information by voice in consideration of the difficulty of hearing by a user, and recording an information output program. Related to the recording medium.
[0002]
[Prior art]
In an information output device that outputs information to a user by voice, if the user hears the information and finds it difficult to hear the information, conventionally, the user inputs that the user could not hear the information, and again outputs the same information repeatedly by voice. The user is asked to listen repeatedly a plurality of times or to improve the quality of the voice, thereby trying to eliminate the difficulty of hearing the voice.
[0003]
[Problems to be solved by the invention]
In the conventional method described above, if the user cannot hear the information, he / she inputs the information.If the user misses the information, the user himself / herself misunderstands the information. There is also a problem that the user cannot be notified and cannot input that the information cannot be heard, so that wrong information is obtained.
[0004]
Also, even if the user inputs that the information could not be heard, the same information is repeatedly output to the user in the same manner again and again. There is a problem that may not be heard again.
[0005]
Furthermore, when the same information is output again and again in the same manner, the user has to listen carefully, and thus there is a problem that the mental burden on the user is large.
[0006]
In addition, when improving the voice quality to eliminate the inaudibility, it depends on the voice quality and the environment in which it is output. There is a problem that you can not.
[0007]
The present invention has been made in view of the above, and it is an object of the present invention to provide an information output apparatus and method and an information output program which have eliminated the difficulty of listening to voice output by adding a description to information that is difficult to hear. An object of the present invention is to provide a recorded recording medium.
[0008]
[Means for Solving the Problems]
In order to achieve the above object, the present invention according to claim 1 is an information output device for outputting information to a user by voice, wherein a word in information to be voice-outputted has a target kana character string of 2 characters. Is decomposed for each kana character and compared. If there is not more than one different kana character in the same consonant string or vowel between the two kana character strings, it is determined that it is difficult to hear. Easily audible judgment means for judging that the word is easy to hear, and additional information for acquiring an additional explanation for eliminating misunderstanding of the word when it is judged that the word is difficult for the user to hear. A voice output for outputting the voice along with the information to be voiced by adding the additional description to the word determined to be difficult to hear by the description obtaining means and the ease of hearing determination means. And summarized in that a means.
[0009]
According to the first aspect of the present invention , for a word in the information to be output as speech, two of the target kana character strings are decomposed for each kana character and compared. If the same consonant string or vowel and one or less different kana characters are present, it is determined that it is difficult to hear, otherwise it is determined that it is easy to hear, and if it is determined that a word is difficult for the user to hear, Since an additional explanation for eliminating the misunderstanding of the word is obtained and the additional explanation is output as speech along with the information to be output as speech, it is possible to eliminate misunderstanding even for words that are difficult for the user to hear.
[0010]
According to a second aspect of the present invention, in the first aspect of the invention, there is provided a target information database storing information requested by a user, input means for receiving an information request from the user, and input means for receiving the information request from the user. Further comprising: an object information search means for searching the object information database for information requested by the requested information; and an additional explanation database storing an additional explanation for eliminating misspelled words. Means is that the means has an additional explanation search means for searching the additional explanation database for the word that is difficult for a user to hear and acquiring the additional explanation for the word.
[0011]
According to the second aspect of the present invention, an information request from a user is received, the information requested by the received information request is searched from a target information database, and an addition for eliminating a word miss mistake is performed. The explanation is stored in the additional explanation database, and the additional explanation database is searched for a word that is difficult for the user to hear, and an additional explanation for the word is obtained.
[0012]
According to a third aspect of the present invention, in the first or second aspect of the present invention, the audibility determining means determines whether or not the word in the information to be output by voice is audible by the voice output. The intelligibility judgment information database storing the judgment information, and searching the intelligibility judgment information database for words in the information output by voice, determine whether the word is easy to hear by voice output. The gist of the present invention is to have a search judging means for acquiring judgment information and judging easiness of hearing the word based on the judgment information.
[0013]
According to the third aspect of the present invention, judgment information as to whether or not the word is easy to hear is stored in the easy-to-hear judgment information database, and the easy-to-hear judgment is made based on the word in the information output as voice. The information database is searched to obtain judgment information on whether it is easy to hear or not, and the easiness of hearing the word is determined based on the judgment information. By providing an additional explanation for the word, it is possible to make it easier to hear.
[0014]
According to a fourth aspect of the present invention, in the first or the third aspect of the present invention, the additional explanation acquiring means replaces a word which is determined to be difficult for the user to hear by the easiness of hearing determination means. Searching the different expression database which stores the paraphrased different expression, and the different expression database with a word determined to be difficult to hear by the easiness of hearing determination means, and obtains another expression for the word as the additional explanation. The gist of the present invention is to have another expression search means.
[0015]
According to the fourth aspect of the present invention, for a word determined to be difficult to hear, another paraphrase of the word is stored in another expression database, and a different word is determined for the word determined to be difficult to hear. In order to search the expression database and acquire another expression for the word as an additional explanation, it is possible to eliminate the difficulty of hearing by adding another paraphrased expression to the hard-to-hear word.
[0016]
According to a fifth aspect of the present invention, there is provided an information output apparatus for outputting information to a user by voice, comprising: a target information database storing information desired by the user; and an input receiving an information request from the user. Means, object information search means for searching the object information database for information desired by the user in response to the information request received by the input means, and information among a plurality of search candidates obtained as search results which are output as sounds. , Two of the target kana character strings are decomposed for each kana character, compared, and the same consonant string or vowel and two or less different kana characters exist between the two kana character strings. In this case, it is determined that it is difficult to hear, and otherwise, it is easy to hear. An additional explanation acquiring means for acquiring an additional explanation for suppressing the misunderstanding of the candidate; and a plurality of search candidates obtained by the object information searching means, sequentially by voice output together with the additional explanation acquired by the additional explanation acquiring means. A candidate enumeration checking means for enumerating and confirming a search target to the user, and providing the user with information desired by the user with respect to the candidate confirmed by the candidate enumeration checking means to be the target of the user. An output means for outputting the data is provided.
[0017]
According to the fifth aspect of the present invention, an information request from a user is received, information desired by the user is searched from the target information database by the received information request, and a plurality of information obtained as a result of the search is obtained. For words in the information to be output as candidates, two of the target kana character strings are decomposed for each kana character, compared, and the same consonant string or vowel is different between the two kana character strings. If there is not more than one kana character, it is determined that it is difficult to hear, otherwise it is determined that it is easy to hear, and for candidates determined to be difficult to hear, an additional explanation that suppresses mistakes in hearing the candidate is obtained, With the additional information, the plurality of search candidates are sequentially enumerated by voice output, a search target is confirmed to the user, and information desired by the user is output for the candidate confirmed by the user. , It is possible to suppress the mistake to hear.
[0018]
According to a sixth aspect of the present invention, in the invention of the fifth aspect, the additional explanation obtaining means is configured to erroneously listen to a candidate determined to be difficult for the user to hear by the easiness of hearing determination means. An additional explanation database that stores an additional explanation that suppresses the above, and a search unit that searches the additional explanation database for a candidate that is determined to be difficult to hear by the easiness of hearing determination unit and obtains an additional explanation for the candidate. That is the gist.
[0019]
According to the sixth aspect of the present invention, for a candidate determined to be difficult to hear, an additional explanation that suppresses mistakes in the candidate is stored in an additional explanation database, and the candidate determined to be difficult to hear is stored. In order to obtain the additional explanation for the candidate by searching the additional explanation database, the search candidates are sequentially enumerated by voice output together with the additional information, the user is checked for the search target, and the candidate confirmed by the user is checked. By outputting information desired by the user, listening mistakes can be suppressed.
[0020]
The present invention according to claim 7 is an information output method for outputting information to a user by voice, wherein an information request from the user is received, and the information desired by the user is received by the received information request. When searching for a desired target from a user by enumerating a plurality of search candidates obtained as a search result by voice output and confirming the target desired by the user, two target kana character strings are decomposed for each kana character. If there is not more than one different kana character that is the same consonant string or vowel between two kana character strings, it is determined that it is difficult to hear. Otherwise, it is determined that it is easy to hear. For the search candidate determined to be, an additional explanation for suppressing misunderstanding of the candidate is obtained, and a plurality of search candidates obtained as the search result are output together with the additional explanation by voice output. The following enumeration, confirms the search to the user, and be required to output the desired information of the user to the user for the user candidate that is the subject has been confirmed for the purpose of.
[0021]
According to the present invention described in claim 7, an information request from a user is received, information desired by the user is searched from the target information database by the received information request, and a plurality of information obtained as a result of the search is obtained. When listing the search candidates by voice output and confirming the target desired by the user, two of the target kana character strings are decomposed for each kana character, compared, and the same between the two kana character strings. If it is a consonant string or a vowel and there is one or less different kana characters, it is determined that it is difficult to hear. Otherwise, it is determined that it is easy to hear. Acquisition of an additional explanation for suppression, a plurality of search candidates are sequentially enumerated along with the additional information by voice output, the user is checked for a search target, and desired information is obtained for the confirmed candidate. The order to output to the user, it is possible to suppress the mistake to hear.
[0022]
The present invention according to claim 8 is a recording medium that records an information output program for outputting information to a user by voice, receives an information request from the user, and requests the user according to the received information request. Information is searched from the target information database, and a plurality of search candidates obtained as a result of the search are enumerated by voice output to confirm a desired target for the user. It is decomposed for each character and compared. If two or more kana character strings have the same consonant string or vowel and one or less different kana character exists, it is judged that it is difficult to hear. Otherwise, it is judged that it is easy to hear. Then, for the search candidate determined to be difficult to hear, an additional explanation for suppressing a mistake in hearing the candidate is obtained, and a plurality of search candidates obtained as the search result are added to the search candidate. Summarize by voice output sequentially with the explanation, confirm the search target to the user, and output the information desired by the user to the user regarding the candidate confirmed to be the target of the user And
[0023]
According to the present invention, an information request from a user is received, information desired by the user is searched from the target information database by the received information request, and a plurality of information obtained as a result of the search is obtained. When listing the search candidates by voice output and confirming the target desired by the user, two of the target kana character strings are decomposed for each kana character, compared, and the same between the two kana character strings. If it is a consonant string or a vowel and there is one or less different kana characters, it is determined that it is difficult to hear. Otherwise, it is determined that it is easy to hear. Acquisition of an additional explanation for suppression, a plurality of search candidates are sequentially enumerated along with the additional information by voice output, the user is checked for a search target, and desired information is obtained for the confirmed candidate. The order recorded in the recording medium information output program for outputting to the user, using the recording medium, it is possible to enhance the flow properties.
[0028]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration of an information output device according to an embodiment of the present invention. In the figure, 1 is an input unit for receiving an input from a user, 2 is a DB search unit for searching a search target desired by the user from information input by the user obtained by the input unit 1, and 3 is The target information database (DB) 4 containing the search target desired by the user is used to confirm the user's desired search target candidates searched by the DB search unit 2 by voice output. The audibility judgment unit 5 for judging whether or not the candidate is easy to hear obtains explanatory information for suppressing the occurrence of a mistake in hearing for the candidate determined to be difficult to hear by the easiness of judgment unit 4. An explanation information acquiring unit 6 is an explanation information database (DB) containing explanation information for suppressing occurrence of a listening error of the search object, and 7 is a search object candidate desired by the user searched by the DB search unit 2. To hear Candidates determined to be difficult to hear by the ease determination unit 4 are listed by voice output to the user together with the explanation information acquired by the explanation information acquisition unit 5 to confirm the search target desired by the user. The confirmation unit 8 is an output unit that outputs information on the search target that has been confirmed by the candidate enumeration check unit 7 as the search target desired by the user.
[0029]
Next, the operation of the information output device configured as shown in FIG. 1 will be described with reference to the flowchart shown in FIG.
[0030]
In step S1, the input unit 1 receives an input of an information request from a user. The DB search unit 2 searches the target information DB 3 based on the information input by the user obtained in step S1, and obtains a search target candidate desired by the user (step S2). The audibility determiner 4 enumerates and confirms the search target candidates desired by the user obtained in step S2 to the user by voice output, and determines whether the candidates are audible or not by the number of characters between the candidates. , Vowels, and consonants, the audibility is determined based on the audibility evaluation function (step S3).
[0031]
The explanation information acquiring unit 5 determines, based on the determination as to whether or not the search target candidate desired by the user obtained in step S3 is easy to hear, for a candidate determined to be difficult to hear, the occurrence of a mistake in hearing the candidate. The explanation information to be suppressed is obtained from the explanation information DB 6 (step S4).
[0032]
The candidate enumeration check unit 7 searches the candidates obtained by the user in step S2 for the search target desired by the user in step S3 and determines that the candidates are difficult to hear in step S3. Is attached to the user with the explanation information obtained in step S4, the user is enumerated and confirmed by voice output, and a search target candidate desired by the user is confirmed (step S5). The output unit 8 outputs the information of the search target, which is confirmed to be the search target desired by the user in step S5, to the user (step S6).
[0033]
Next, the operation of the information output device shown in FIG. 1 will be specifically described with reference to FIGS. FIG. 3 is a diagram showing an example of a use situation used for explaining a specific example. Specifically, a search target "address: Yokohama city, Kanagawa prefecture, OO town x chome, name: Toru Sato ( Yomi: Satotoru) ”, information“ telephone number ”that the user wants to know about the search object, and information“ address: Yokohama city, 名 town, name: Sato ”input by the user in the input unit 1. FIG. 4 is a DB search. FIG. 5 shows an example of an evaluation function used in the audibility judgment unit 4 for judging audibility, and FIG. 6 shows an audibility judgment unit used in the object information DB 3 used in the unit 2. 7 is an example of the contents of the explanation information DB used by the explanation information acquisition unit 5, and FIG. 8 is a candidate enumeration confirmed by the candidate enumeration confirmation unit 7 by voice output. It is a figure which shows the example of an enumeration of.
[0034]
First, as shown in FIG. 3, the user wants to know the information (the telephone number of the search target) about the search target of “address: XX town x chome, Yokohama city, Kanagawa prefecture, name: Toru Sato (reading: Sato Toul)”. In order to search the stored target information DB 3, the user inputs “address: Yokohama city, ○ town, name: Sato” into the input unit 1.
[0035]
The DB search unit 2 searches the target information DB 3 using the “address: Yokohama city, 町 town name: Sato” received by the input unit 1. Since the contents of the target information DB 3 are as shown in FIG. 4, as a candidate of a search target desired by the user corresponding to “address: Yokohama city, 町 town name: Sato”, “address: Yokohama city, Kanagawa prefecture, ○ town name” : Toru Sato (read: Satou Toru) ",
"Address: Yokohama, Kanagawa Prefecture, XX town name: Toru Sato (read: Sato Toru)",
"Address: Yokohama, Kanagawa Prefecture, ○○ Town name: Kaoru Sato (read: Sato Kaoru)"
Is obtained.
[0036]
In the audibility judgment unit 4, the search target candidates obtained in the DB search unit 2 are "address: Yokohama city, Kanagawa prefecture, ○ town name: Toru Sato (read: Sato Toul)",
"Address: Yokohama, Kanagawa Prefecture, XX town name: Toru Sato (read: Sato Toru)",
"Address: Yokohama, Kanagawa Prefecture, ○○ Town name: Kaoru Sato (read: Sato Kaoru)"
Of the user is enumerated and confirmed by voice output. Here, the evaluation function F shown in FIG. 5 is used as a determination method.
[0037]
This evaluation function F takes two kana character strings as variables. The two kana character strings are decomposed for each kana character, compared, and if the same consonant string or vowel and one or less different kana characters exist between the two kana character strings, the evaluation function F is 0. return it. That is, when the value of the evaluation function F is 0, even if two kana character strings are enumerated and confirmed to the user by voice output, the user determines that the two kana character strings are easy to miss. FIG. 6 shows a result of the judgment of the easiness of hearing the searched candidate by the evaluation function F.
[0038]
The explanation information acquiring unit 5 acquires, from the explanation information DB 6, explanation information that resolves the difficulty of hearing for a candidate determined to be difficult to hear (that is, easy to miss) by the ease of listening determination unit 4. Candidates determined to be difficult to hear by the easiness of hearing determination unit 4 are:
"Address: Yokohama, Kanagawa Prefecture Name of the town: Toru Sato (read: Sato Toul)"
And "Address: Yokohama, Kanagawa Prefecture Name of town: Toru Sato (read: Sato Toru)"
Therefore, the explanation information is obtained for both of them.
[0039]
Since the contents of the explanation information DB 6 are as shown in FIG. 7, for the candidate “address: ○ town, Yokohama City, Kanagawa Prefecture, name: Toru Sato (reading: Sato Toul)”, the explanation information: “Middle is pronounced as Aio of Ueio. For the candidate "Address: Yokohama City, Kanagawa Prefecture, XX Town Name: Toru Sato (read: Sato Toru)", explanation information: "Difficult to hear" To Toru, which reads "Toru in the main street". It is acquired as explanation information to be eliminated (that is, to suppress listening mistakes).
[0040]
In the candidate enumeration checking unit 7, the search target candidates desired by the user searched in the DB search unit 2 are obtained by the explanation information obtaining unit 5 for the candidates determined to be difficult to hear by the easiness of hearing determination unit 4. Along with the explanation information, the information is enumerated by voice output to the user to confirm the search target desired by the user (FIG. 8).
[0041]
The output unit 8 outputs information on the search target that has been confirmed by the candidate enumeration check unit 7 as the search target desired by the user. That is, as shown in FIG. 8, the candidate enumeration confirmation unit 7 confirms that the candidate is the search target desired by the user, “Address: Yokohama City, Kanagawa Prefecture, No. Name: Toru Sato (read: Sato Toul)” Output the phone number to the user.
[0042]
By the above processing, candidates as shown in FIG. 4 are searched as the search target candidates desired by the user, and the mere confirmation of the enumeration by voice output, such as “Toul” and “Toru,” easily causes the user to misunderstand. Even in this case, the audibility of the candidate to be enumerated and confirmed is determined, and as shown in FIG. 8, by attaching explanatory information to the candidate to be enumerated, occurrence of erroneous listening is eliminated, and the search target desired by the user is determined. Can be enumerated and confirmed by voice output, and information on the search target desired by the confirmed user can be output.
[0043]
【The invention's effect】
As described above, according to the present invention, before a user listens to information, it is determined whether or not a word in information output as voice is easy for the user to hear. When it is determined that the word is difficult to hear, an additional explanation for eliminating the misunderstanding of the word is obtained, and this additional explanation is output as voice along with information to be output by voice. , Can be heard accurately. In addition, even if the voice quality is poor, it is possible to eliminate a user's mistake in listening, and it is also possible to eliminate a mistake in listening due to the user's own assumption.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of an information output device according to an embodiment of the present invention.
FIG. 2 is a flowchart showing the operation of the information output device shown in FIG.
FIG. 3 is a diagram illustrating an example of a usage status used for describing a specific example of the embodiment in FIG. 1;
FIG. 4 is a diagram showing an example of contents stored in a target information database used in the information output device of FIG. 1;
FIG. 5 is an explanatory diagram showing an example of an intelligibility evaluation function used by the intelligibility determination unit used in the information output device of FIG. 1 to determine the intelligibility.
FIG. 6 is an explanatory diagram showing an example of candidates to be enumerated and confirmed by voice output determined by a audibility determining unit used in the information output device of FIG. 1;
FIG. 7 is a diagram showing an example of contents stored in an explanation information database used in the information output device of FIG. 1;
8 is an explanatory diagram showing an example of an enumeration method by voice output to a user performed by a candidate enumeration confirmation unit used in the information output device of FIG. 1;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Input part 2 DB search part 3 Object information database 4 Easiness of hearing judgment part 5 Explanation information acquisition part 6 Explanation information database 7 Candidate enumeration confirmation part 8 Output part

Claims (8)

An information output device that outputs information to a user by voice,
For words in the information to be output as voice, two of the target kana character strings are decomposed for each kana character and compared, and between the two kana character strings, the same consonant string or vowel and a different kana character If there is one or less, it is determined that it is difficult to hear, otherwise, it is easy to hear.
An additional explanation acquiring means for acquiring an additional explanation for eliminating a misunderstanding of the word when it is judged that the word is difficult for the user to hear by the audibility judgment means;
An information output device, comprising: voice output means for adding the additional explanation to the word determined to be difficult to hear by the ease of hearing determination means and outputting the word together with information to be output as voice. A target information database storing information requested by the user, input means for receiving an information request from the user, and a target information search for searching the target information database for information requested by the information request received by the input means Means, and an additional explanation database storing additional explanations for eliminating misspellings of words.
2. The information output apparatus according to claim 1, wherein the additional explanation obtaining unit includes an additional explanation search unit that searches the additional explanation database for the word that is difficult for a user to hear, and acquires the additional explanation for the word. apparatus. The audibility determination means includes a audibility determination information database that stores determination information as to whether or not the word in the information output as voice is audible by voice output, and a voice output. Searching the intelligibility judgment information database with a word in the information to obtain judgment information as to whether the word is easy to hear by voice output, and based on the judgment information, determine the ease of hearing of the word. 3. The information output device according to claim 1, further comprising a search determination unit for determining. The additional explanation obtaining means includes, for a word determined to be difficult for the user to hear by the ease of hearing determination means, another expression database storing another paraphrase of the word; 4. The apparatus according to claim 1 , further comprising a different expression search unit that searches the different expression database for a word determined to be difficult to hear by the determination unit, and acquires another expression for the word as the additional explanation. Information output device. An information output device that outputs information to a user by voice,
A target information database storing information desired by the user,
Input means for receiving an information request from a user;
Target information search means for searching the target information database for information desired by the user according to the information request received by the input means;
For words in the information output as a plurality of search candidates obtained as search results, two of the target kana character strings are decomposed for each kana character, compared, and between the two kana character strings. If the same consonant string or vowel and at least one different kana character is present, it is determined that it is difficult to hear, and if not, it is easy to hear.
Additional explanation acquiring means for acquiring an additional explanation for suppressing misunderstanding of the candidate for a search candidate determined to be difficult to hear by the easiness of listening determination means,
Candidate enumeration checking means for sequentially listing a plurality of search candidates obtained by the target information search means by voice output together with the additional explanation acquired by the additional explanation acquisition means, and confirming the search object to the user;
Output means for outputting to the user information desired by the user with respect to the candidate that has been confirmed to be the target of the user by the candidate enumeration checking means. The additional explanation obtaining means includes an additional explanation database storing an additional explanation for suppressing a mistake in listening to the candidate for a candidate determined to be difficult for the user to hear by the ease of listening determination means; searches the additional description database is determined pile interviews with ease determination unit candidate, the information output device according to claim 5, characterized in that it comprises a retrieval means for obtaining an additional explanation for the candidate. An information output method for outputting information to a user by voice,
Receiving information requests from users,
Based on the received information request, the user searches the target information database for desired information,
When a plurality of search candidates obtained as a result of the search are enumerated by voice output to confirm a target desired by the user, two target kana character strings are decomposed for each kana character, compared, If the same consonant string or vowel and two or less different kana characters exist between two kana character strings, it is determined that it is difficult to hear, otherwise it is determined that it is easy to hear,
For the search candidate determined to be difficult to hear, an additional explanation for suppressing the mistake of hearing the candidate is obtained,
A plurality of search candidates obtained as the search result are sequentially enumerated by voice output together with the additional explanation, and the user is checked for a search target,
An information output method characterized by outputting information desired by a user to a user for a candidate confirmed to be a target intended by the user. A recording medium recording an information output program for outputting information to a user by voice,
Receiving information requests from users,
Based on the received information request, the user searches the target information database for desired information,
When a plurality of search candidates obtained as a result of the search are enumerated by voice output to confirm a target desired by the user, two target kana character strings are decomposed for each kana character, compared, If the same consonant string or vowel and two or less different kana characters exist between two kana character strings, it is determined that it is difficult to hear, otherwise it is determined that it is easy to hear,
For the search candidate determined to be difficult to hear, an additional explanation for suppressing the mistake of hearing the candidate is obtained,
A plurality of search candidates obtained as the search result are sequentially enumerated by voice output together with the additional explanation, and the user is checked for a search target,
A recording medium recording an information output program for outputting information desired by a user to a user for a candidate confirmed to be a target intended by the user.

Images