JPH04200050A - Telephone voice reply device - Google Patents

Abstract

PURPOSE:To easily confirm the user and understand content of information by converting a character data obtained through the retrieval of a prescribed character data memory into a voice data according to a numeral data specified by an individual code and outputting the data to a telephone line as a reply. CONSTITUTION:An individual code obtained through identification and recognition processing by a PB identification/voice recognition section 2 is used for the retrieval of an individual code memory 3 and when a relevant individual code is registered, an individual character data memory 4 is retrieved according to the information and the registered character data is read corresponding to the individual code. The character data read from the individual data memory 4 is given to a voice rule synthesis section 6, in which the character data is converted into a voice data. Since the character data retrieved and generated based on a numeral data specified by the individual code is voice-synthesized and outputted, the understanding of the message content subjected to voice reply is easily understood.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application) The present invention relates to a telephone voice response device that outputs information inquired via a telephone line in an easy-to-understand voice.

(Prior Art) Recently, various voice response services have been provided through telephone lines. For example, various system capabilities have been developed to provide telephone voice response services such as balance inquiries in banking and event information in department stores.

This type of system inputs predetermined information by using a DTMF signal (Busoshuhon signal) through the nearest telephone line, or by inputting predetermined information by telephone voice. The response to the request is obtained by telephone voice. In this case, the user inputs an individual code specified for each user in advance, identifies the user according to the input individual code, and then outputs information about the user by voice. will be held.

Generally, the system side confirms the individual code input by the user by outputting the individual code audibly. For example, as an individual code [1234]
If a numeric string is entered, the system will convert it to
)is not it. ], etc., to prompt confirmation.

However, there is a problem in that it is very troublesome to confirm this type of number string by outputting it audibly. In particular, if only audible digits are output as response information, there is a problem in that the information is not transmitted sufficiently.

(Problem to be solved by the invention) In this way, in the conventional telephone voice response device, when numeric data is given, it simply converts it into voice data and outputs the response. There were some problems, such as making it difficult to understand.

The present invention has been made in consideration of these circumstances, and its purpose is to provide a telephone voice response device capable of outputting a voice response that is easy for the user to confirm and understand the information content. It is about providing.

[Structure of the Invention] (Means for Solving the Problems) A telephone voice response device according to the present invention recognizes a DTM F signal or numeric voice input from a telephone line, and recognizes the individual code recognized by this means. In addition to outputting the character data corresponding to the above as voice data, the character data obtained by searching a predetermined character data memory is converted into voice data by voice rule synthesis according to the numerical data specified by the above-mentioned individual code, and the phone line is connected. This feature is characterized in that it outputs a response to.

(Function) According to the present invention having such a function, character data corresponding to an individual code given as numeric data is obtained, this character data is converted to voice data, and the response is output to a telephone line. Compared to the case where the individual code itself is converted into voice data and output as a response, it becomes possible to confirm the information content more easily.

Further, since the character data generated from the numerical data specified by the individual code is converted into voice data by voice rule synthesis and output, the information content can be easily understood.

(Example) Hereinafter, a telephone voice response device according to an example of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing the basic schematic configuration of the embodiment device.
is the network control unit (NCU). The embodiment device has this N
Connected to the telephone line via CUI.

The PB identification/speech recognition unit 2 receives the NCUI from the telephone line.
DTMF signal input via
It identifies and recognizes PB multiplied signals and audio signals. In this PB identification/speech recognition unit 2, the input DTM
The F signal is identified and the numerical data indicated by the DTMF signal is determined, or the numerical data input as voice information is recognized and processed. Specifically, the PB identification/speech recognition unit 2 identifies and recognizes the individual code indicated by numerical information input by the user via the telephone line.

On the other hand, in the individual code memory 3, individual codes for each user set in advance by the user are registered, and in the individual character data memory 4, the user's name, etc. is registered in correspondence with each individual code. Character data is registered. Registration and setting of data in the individual code memory 3 and the individual character data memory 4 is performed via the data input section 5.

The individual code obtained through identification and recognition processing in the PB identification/speech recognition section 2 is used to search the individual court memory 3, and if the corresponding individual code is registered, it is searched according to the information. The individual character data memory 4 is searched and the character data registered corresponding to the individual code is read out. The character data read from the individual character data memory 4 is given to the voice rule synthesis section 6, where the character data is converted into voice data.

The speech rule synthesis section 6, for example, as shown in FIG. Phonological parameter string generation section 14. Prosodic parameter string generation section 15. It is configured to include a speech synthesis section 16.

The character string analysis section 12 analyzes the input character string with reference to the accent dictionary 11 to obtain its phonetic symbol string and prosodic symbol string. Phonological parameter string generation unit 14
The system inputs a phoneme symbol string corresponding to the input character string obtained by the character string analysis section 12, and generates a phoneme parameter string to be used for speech rule synthesis by referring to the speech segment file 13. . At this time, the prosodic parameter string generation section 15 generates a prosodic parameter string to be used for speech synthesis according to the prosodic symbol string for the input character string determined by the character string analysis section 12. The speech synthesis unit 16 generates a sound source according to the phonological parameter string and the prosodic parameter string, performs modern digital filtering processing on its vocal tract characteristics, and synthesizes the speech indicated by the input character string according to predetermined rules. It is something to do.

The synthesized speech generated by the speech rule synthesis section 6 is output as a speech message to the telephone line via the NCUI.

In addition, instead of obtaining the character data corresponding to the individual code directly from the individual character data memory 4 as described above,
For example, as shown in FIG. 3, a numeric data memory 7 storing numeric data required for each user is searched using the individual data, and the numeric data retrieved from this numeric data memory 7 is transferred to a character data generating section 8. The apparatus may be configured so that the data is converted into character data and provided to the speech rule synthesis section 6.

Furthermore, as shown in FIG. 4, the numeric data stored in the numeric data memory 7 is calculated in a spreadsheet 9, and the spreadsheet result is converted into character data in the character data generator 8. The apparatus may be configured so that the information is provided to the speech rule synthesis section 6.

That is, the telephone voice response device according to the present invention does not directly convert an individual code indicated by numerical information input via a telephone line into voice and output a response, but instead sets the individual code in correspondence with the individual code. The present invention is characterized in that character data such as a person's name, etc., or character data generated according to numeric data given corresponding to an individual code is synthesized into voice and output as a response via the telephone line.

The specific processing operation of the embodiment apparatus configured as described above will be explained by taking as an example the case where it is applied to a test score adjustment system.

Test result data is input in advance through the data input section 5, and is registered in the character data memory 4 and numerical data memory 7 in correspondence with the examinee's individual code. This data input may be, for example, your student ID number.

A password, a name, a score in Japanese, a score in mathematics, and a score in English are each input as information on each item.

If information on the overall ranking is also registered, it can be obtained by, for example, activating the spreadsheet section 9 and executing the ranking calculation.

When data input settings are performed in this manner, the system displays a title screen as shown in FIG. 5, for example. Icon (Fl) on this title screen
By selecting the icon (F2), a state of waiting for an incoming call for test results is set, and by selecting the icon (F2), a mode for changing score data is set.

When the icon (Fl) is selected, the display screen is changed as shown in FIG. 6, and incoming call reception is turned on. In this state, if there is an incoming call via the telephone line, this incoming call will be detected and the voice rule synthesizer 6 will be activated, for example, "This is O○ Preparatory School/Examination Information Service Center. Please give me your student ID number. A beep voice message and timing sound are output in response. In response to such a response message, the user inputs, for example, a four-digit student ID number and [#] by operating the dial pad of the telephone, and inputs this as a D ThIF signal. Incidentally, it is also possible to input the student ID number by voice of numbers by changing this D T MF multiplication signal.

The student ID number input in this way is identified and recognized by the PB identification/speech recognition section 2, and the individual code memory 3 is searched according to the identification/recognition result. On this occasion,
If the corresponding student ID number is not registered in Individual Court Memory 3, a message such as "Your student ID number is incorrect. Please try again.

A "beep" response is emitted from the voice rule synthesizer 6. After the student ID number has been confirmed, the user is then prompted to input a password, and the individual coat memory 3 is similarly searched for this password and the confirmation process is performed.

When it is confirmed that the user is a makeshift person or a regular user through the above confirmation process, the data registered about the user is pressed into response as a voice message.

A voice message response to this arrangement is output by searching the character data memory 4 or numeric data memory 7 based on the individual code to find the corresponding data, and generating predetermined character data in accordance with the data.

Specifically, do not output individual codes as they are (,
The name information corresponding to the individual code is obtained as character data, and in addition to this name information, the data of the item specified by the above individual code in the data shown in Figure 6 is read out, and this is sent to the predetermined message. It generates character data converted into a format, synthesizes these character data according to phonetic rules, and outputs the result.

For example, student ID number C, 9001], password number [2020]
In response to this input, the user's [calculating Katoichirouko, Japanese language score and its evaluation C70/B], mathematics score and its evaluation [C70/B] from the data shown in FIG. 85/Al, English score and evaluation [67/
B] and the overall ranking [10]. And, as a combination of these data, ``The rank of Katoichirosan is 10th out of x groups.

70 points in Japanese, 85 points in mathematics, and 67 points in English.

I'm good at mathematics. ” is generated, synthesized into speech, and output.

In addition, the underlined part in the above character data is the individual data memory 3.
and a message portion generated based on these information. By outputting the character data corresponding to the individual code entered in this way, in this example the user's name, as a voice message, the information is extremely easy to hear and does not contain any strange information such as a PIN number. It can be used as a content.

Regarding message generation based on test data,
For example, the data for each test item is ranked according to its score as 0-39...C, 40-79...B, 80-100...A, etc., as shown in Figure 7. This is done by selectively finding a corresponding response message from a table in which response messages are registered and set in advance according to each combination of ranks. Therefore, in terms of character data generation processing, for example, ■ it has excellent results.

■　Excellent grades, but XX or not good.

■　XX is excellent.

■ A little more effort is required.

■ It requires a lot of effort.

■ Does it require effort? XX is excellent.

Based on the basic pattern of [
Character data forming a voice message may be generated by adding information of the corresponding item to the XX portion.

Thus, when outputting the information specified by the student ID number or PIN as a voice response as described above, instead of the student ID number or PIN, the name corresponding to the student ID or PIN is used as voice data, and furthermore, the number According to this device that outputs character data generated based on data as voice data, the content of the response information can be Able to communicate accurately and clearly.

Therefore, great practical effects can be achieved, such as being able to respond very well to telephone inquiries.

Note that the present invention is not limited to the embodiments described above. For example, the spreadsheet unit 9 may calculate the ranking of scores for each item, the average score, the deviation value, etc., and output these as a response as a voice message. Moreover, it can be applied not only to inquiries about test result data but also to other telephone arrangement systems. In short, the present invention can be implemented with various modifications without departing from the gist thereof.

[Effects of the Invention] As explained above, according to the present invention, the character data corresponding to the individual code inputted by the user through the telephone line is synthesized and output as voice, and the numeric data specified by the individual code is outputted as voice. Since the text data searched and generated based on the text is synthesized into speech and output, it has great practical effects, such as making it easier to understand the contents of the message that is answered by voice.

[Brief explanation of the drawing]

The figures show a telephone voice response device according to an embodiment of the present invention; FIG. 1 is a diagram showing a basic configuration example of the embodiment device, and FIG. 2 is a voice rule synthesis unit used in the embodiment device. Figures 3 and 4 are diagrams each showing a specific example of the configuration of the embodiment device, and Figures 5 to 7 are diagrams in which the present invention is applied to an inquiry system for test result data. FIG. 3 is a diagram for explaining an example. DESCRIPTION OF SYMBOLS 1...Network control unit (NCU), 2...PB identification/speech recognition unit, 3...Individual code memory, 4...Character data memory, 5--Data input unit, 6-Speech rule synthesis unit ,7
...Numeric data memory, 8...Character data generation section,
9...Spreadsheet section. Applicant's representative Patent attorney Takehiko Suzue 11!1 Figure 2 Figure 3 Figure 4 Figure 5

Claims (1)

[Claims] A means for recognizing a DTMF signal or numeric voice input from a telephone line, and a voice rule for searching a predetermined character data memory based on the numeric data specified by the individual code recognized by this means. 1. A telephone voice response device, comprising means for converting into voice data through synthesis and outputting a response to a telephone line.