JP2002044258A - Telephone voice response device for activating program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a telephone voice response device which facilitates information providing service and retrieval service with voice response only by generating a program for the text input/output. SOLUTION: Key information is obtained by speech recognition, DTMF recognition, etc., in voice response exchange, and an external program is activated by using the key information as an argument. Then a text outputted by the external program is received and the text information is transformed into a voice waveform and returned to a user. The communications with the external program are restricted within the argument at the activation of the external program, the output results and the finish signal to architect the voice service easily.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer having a telephone function for receiving an incoming call from a partner and answering the call. And a telephone voice response device based on a general-purpose computer.

[0002]

2. Description of the Related Art Conventionally, when constructing a telephone voice response service for providing information from text-based data, a telephone voice hardware is operated, information is obtained from a database or the like, and text data is obtained. It is realized by making a program that converts voice from.

[0003]

In the conventional environment described above, information is acquired from a database or the like for a system to be constructed, and in addition to text shaping processing, telephone voice hardware processing, text-to-speech conversion. There was a problem that processing had to be included in the program.

In addition, many programs for processing information such as a database are premised on text input and text output. If these programs can be used for telephone voice services, it is easy to realize a telephone voice response service. become.

An object of the present invention is to provide an apparatus for realizing a voice response service by acquiring information from a database or the like and creating a program for performing text shaping processing.

[0006]

In order to solve the above-mentioned problems, the present invention provides a communication means between the voice response proxy software (2) and the external program (3) by using an argument when starting the external program (3), Means (3) for outputting text and making a signal at the end is used. By the above means,
The external program (3) can create a program with emphasis on text input / output.

The flow of processing of the voice response proxy software (2) will be described with reference to FIG. The voice response proxy software (2) is started (S1), and the telephone voice hardware (1) is operated to monitor whether there is an incoming call. (S
2) When a call is connected, a unique call ID is generated for one call connection. (S3) The voice guidance prompting the input of the key basic information is transmitted.
(S4) Specifically, "It is an employee database. Please enter an employee code.""It is a stock price search database. Please say your company name." Get key basic information. The key basic information is a DTMF signal or utterance. (S5) The key basic information is converted into key information. Key basic information is D
In the case of a TMF signal, a numerical value obtained by DTMF detection is used as key information, and when the key basic information is voice, a character obtained by voice recognition is converted as key information. (S6) The external program (3) is started using the key information and the call ID as arguments. (S7) Wait until the external program ends, and receive a text output result or a predetermined disconnection signal. (S
8, S9) If a disconnect signal is received, the call is disconnected and terminated. (S
12) When receiving the output result, convert the text output result into audio waveform information. (S10) Output voice to the telephone line. (S11) Return to S5 and repeat the above.

[0008] The external program is to be output as text. Here, the reason why the key information is passed to the external program as an argument is that the key information is used when the external program acquires information from a database or the like or as a means for determining a user's operation decision. Here, the reason for passing the call ID as an argument to the external program is that the external program terminates after outputting the text, so if the external program is started many times in the same call connection, what operation has been performed before? This information is used by external programs to manage this information.

When the key basic information is voice, the key information is created by voice recognition. The voice recognition refers to a general technique of analyzing a voice waveform and generating text information from a voice waveform pattern. Point. In addition, speech waveform information is generated from text information. For this purpose, a general technique is used in which text is analyzed and converted into phonograms, a speech waveform corresponding to the phonogram is synthesized, and speech waveform information is generated. .

[0010] Since communication with an external program is made up of only an argument at the time of startup, a text output at the time of termination, and a signal of termination, a voice service can be created simply by creating an external program only for text processing without being aware of a voice application. Can start.

In addition, the apparatus according to claim 2 obtains a destination number from a signal for controlling call processing when receiving a call, and stores the destination number in a storage unit (5) in which a telephone number is associated with a program. The caller's number is compared with the telephone number in the storage unit, and a program corresponding to the matched telephone number is determined as an external program (3) to be activated by the voice response proxy software (2).

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The configuration of the apparatus (4) is such that telephone voice hardware (1) is attached to computer hardware and controlled by voice proxy software (2). The telephone voice hardware (1) here refers to hardware that has an interface connected to a telephone line, performs a call connection procedure, and digitally processes voice. The device has an interface for IP connection with a telephone line interface. A general-purpose OS is mounted on the computer hardware, and the voice proxy software operates on the OS.

In order to implement a voice service in the device, it is necessary to first create an external program (3).
The external program (3) is created by a system administrator, and also operates on the OS. It is assumed that the external program (3) is designed to be activated by using the information key and the call ID as arguments, acquire information from the electronic information database, format the text, and output the text.

A storage unit (5) in the apparatus according to claim 2
Is used. The storage unit (5) may be inside or outside the device.

In the above embodiment, a specific example of the external program will be described with reference to FIG. The external program (3) is created in advance so as to output a text indicating the stock price of the company name when the external program (3) is started with the company name as an argument. The user makes a call from the telephone (8) to the device (4) via the telephone line. The device responds,
"This is a stock search database. Please say your company name." The user utters “A trading”. Then, the device recognizes the voice from the voice waveform received as “A Shoji” and generates characters “Ashoji”. The word "Ashoji" is recognized as key information, and the external program (3) is started with the argument "Ashoji". From the external program (3), "A
The trading price of Shoji is 100 yen. Please say the name of the company. "The text output is obtained. The above text output is obtained, converted into an audio waveform, and the user is told," The stock price of A Shoji is 100 yen. Please say the company name. " And send it out by voice. This operation is repeated as described above. When the external program (3) wants to disconnect the call intentionally, the external program (3) is terminated by an abnormal end signal and the call is disconnected.

The reason why the call ID is not described in the above-mentioned use example is to simply represent an example in which the external program (3) is started using the information key. Usually, the call ID is also passed as an argument at the time of startup, and is used by the external program for the number of startups, authentication, reference to the previous result, and the like. As a result, a flexible voice service can be provided. Further, the above-mentioned use example does not describe the implementation of claim 2.

[0017]

As shown in FIG. 3, when a system such as a schedule table or inventory management is already constructed using a PC client terminal (7) and an electronic information database server (6), the device (4) is used. ) Is installed on a LAN, and a voice response service can be added by creating an external program (3) using a client program used in the PC client terminal (7). Even if the client program does not operate on the device due to a difference in OS or the like, if the electronic information database server (6) is general-purpose and there is a general-purpose client program operating on the device, Using this, a voice response service can be easily started. It is an advantage of the present invention that the voice response service can be easily added as described above.

[Brief description of the drawings]

FIG. 1 is a flowchart of voice proxy software that performs overall processing according to an embodiment.

FIG. 2 is a block diagram of the telephone voice response apparatus according to the embodiment.

FIG. 3 is a diagram illustrating a network configuration according to the embodiment;

FIG. 4 is a diagram showing an entire process including an operation example of an external program in the embodiment;

[Explanation of symbols]

1 is telephone voice hardware, 2 is voice response proxy software, 3 is an external program, 4 is the device concerned, 5 is a storage unit, 6 is an electronic information database server, 7 is a PC client, and 8 is a telephone used by a user. Terminal, S1 to S
12 is a process block of the flowchart.

Claims (2)

[Claims] 1. A general-purpose computer-based telephone voice response apparatus for receiving an incoming call from a destination and exchanging a voice response with the destination, generating a unique value for each call. And accepts a DTMF signal input via a telephone line,
Means for detecting the F signal and converting it to key information, or means for receiving an input of voice via a telephone line, recognizing the voice and converting it to key information, and using the keyword and the unique value as arguments in the apparatus. Means for activating the external program, obtaining a text output of the external program, converting the text output into an audio waveform, and sending the voice output to the destination, detecting an abnormal termination signal of the external program, and A telephone voice response device, comprising: 2. A storage device according to claim 1, further comprising: means for detecting a destination number from a call processing control signal at the time of an incoming call; and a storage unit for storing a predetermined telephone number and an external program to be started in association with each other. Means for retrieving the storage unit from the destination number and changing the external program activated by the destination number.