JP2006093761A - Voice automatic response system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice automatic response system for automatically responding to a voice through a phone call, surely carrying out voice recognition even if a plurality of phone calls arrive at the same time and performing proper transfer processing to the phone calls. <P>SOLUTION: The voice automatic response system is configured to connect parts for controlling ISDN lines to parts for controlling VoIP lines and can transfer a phone call received by an ISDN to a destination via a VoIP. The voice automatic response system is configured such that it acquires a license of a voice recognition engine and occupiedly uses the engine when voice recognition is required by the processing of a voice automatic response, releases the license of the voice recognition engine in the case of line transfer not needing voice recognition, and does not occupy the voice recognition engine. The voice automatic response system manages flags such as a flag denoting a recorded voice part and a flag denoting a synthesized voice part so that the voice response can comprise only a recorded voice, or only a synthesized voice, or a combination of the recorded voice and the synthesized voice. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an automatic voice response system through a telephone, and relates to an automatic voice response system that performs voice recognition reliably even when a plurality of incoming calls are received at the same time and realizes an appropriate telephone transfer process.

  2. Description of the Related Art Conventionally, an automatic voice response system has been devised in which voice recognition is performed by a computer through a telephone and a telephone transfer process desired by a user is automatically performed (for example, see Patent Document 1).

  However, the conventional method is a transfer process that uses only the public line network, occupies the voice recognition engine until the transfer is completed for each call, and the voice response is only recorded voice or synthesized voice only It was difficult to hear and lacked flexibility.

JP 2001-346236 A

  As described above, the conventional techniques have the following problems.

  Conventional automatic voice response systems for computers through telephones use only public lines for telephone transfers, or occupy the voice recognition engine even during line transfer processing that does not require a voice recognition engine. It is efficient, and the voice response uses only the recorded voice or only the synthesized voice, which makes it difficult to hear and lacks flexibility.

  An object of the present invention is to transfer a telephone call made with ISDN (Integrated Services Digital Network) to an Internet telephone using VoIP (Voice over IP) or only when it is really necessary in an automatic voice response system of a computer through a telephone. The voice recognition engine is used to prevent the voice recognition engine from being occupied unnecessarily, and the voice response can be a combination of the recorded voice and the synthesized voice. .

  In order to solve the above problems, the present invention takes the following means.

  A part for controlling the line of ISDN and a part for controlling the line such as VoIP are connected so that a telephone received by ISDN can be transferred to the other party via VoIP.

  In the automatic voice response process, the voice recognition engine license is acquired and used exclusively when voice recognition is required, and the voice recognition engine license is released and occupied when transferring a line that does not require voice recognition. Configure not to.

  By managing the flag of the voice response, such as a flag indicating the recorded voice part and a flag indicating the synthesized voice part, the voice response is made only to the recorded voice, only the synthesized voice, or a combination of the recorded voice and the synthesized voice. Configure to be able to.

  According to the present invention, the following effects can be expected.

  By using VoIP when transferring the telephone line, it is possible to reduce telephone charges, and by occupying the voice recognition engine only when necessary, even when multiple calls are received, Since all voice recognition engines are occupied, it becomes possible to prevent the response from becoming impossible, and by combining the voice response with the recorded voice and synthesized voice, it is possible to respond flexibly as an easy-to-hear voice You will be able to

  The present invention takes the following embodiment.

  The ISDN line control unit and the VoIP control unit are connected.

  As a result, a call received through the ISDN line can be transferred to the other party by an Internet telephone using VoIP.

  A voice recognition engine license is acquired only when voice response with a telephone user is required, and the voice recognition engine is occupied by releasing the license when voice recognition is no longer required, such as when transferring a line. Control to make the time as short as possible.

  As a result, the time occupied by each of the plurality of voice recognition engines can be shortened as much as possible, and an appropriate automatic response can be made even when a plurality of calls are received.

  A voice response is provided with a flag for discriminating a recorded voice portion and a synthesized voice portion, and the flag is managed so that the recorded voice and the synthesized voice can be combined.

  As a result, the fixed portion of the sentence is easily heard by the recorded voice, and only the portion that becomes the variable sentence is the synthesized voice, so that the voice response can be flexibly and easily heard.

  A typical embodiment according to the present invention will be described with reference to the drawings. In the following, the same portions are denoted by the same reference numerals, and detailed description may be omitted.

  FIG. 1 shows an embodiment of the present invention.

  When the user makes a call to the voice automatic answering system, the line control unit 1 investigates the status of the available line and assigns the available line to the voice transmitting / receiving unit 2. The voice transmission / reception unit 2 uses the allocated line to perform a process of designating a transfer destination by the user and voice.

  When performing such voice exchanges, the voice transmission / reception unit 2 acquires a license for the voice recognition engine 3, occupies one of the voice recognition engines 3, passes the voice to the voice recognition engine 3, and analyzes the contents. Is done by doing.

  When this voice recognition processing is completed, it is no longer necessary to occupy the voice recognition engine 3, so the license is released and the voice recognition engine 3 can be used on other lines.

  When the transfer destination is determined in this way, the voice of the user is transferred to the transfer destination by the voice transfer unit 4. At this time, if a VoIP line is available by the server as shown in FIG. However, if all the VoIP lines are in use, the telephone can be transferred from the user as much as possible by transferring the call using the ISDN line.

  When any line is in use, a reply is made that the transfer service is currently busy and cannot be used.

  The transfer destination designation by voice has a flag indicating whether it is a recorded voice or a synthesized voice. By managing this flag, the recorded voice is used in combination with the recorded voice and the synthesized voice. This makes it easy to hear and uses a synthesized voice for variable parts so that it can be flexibly supported.

  By performing the control as described above, as shown in FIG. 3, by calling the automatic voice response system, there is an automatic voice response, and by specifying the transfer destination by voice according to this instruction, The telephone line can be transferred to the target transfer destination.

It is a block diagram of the present invention. It is a server block diagram of this invention. It is a use figure of this system.

Explanation of symbols

1: Line control unit 2: Voice transmission / reception unit 3: Voice recognition engine 4: Voice transfer unit

Claims (3)

In the automatic voice response system that automatically transfers the phone by specifying the transfer destination by voice response with the computer through the phone,
Calls received from ISDN lines are transferred via Internet telephone via VoIP.
An automatic voice response system characterized by that. In the automatic voice response system that automatically transfers the phone by specifying the transfer destination by voice response with the computer through the phone,
Occupies the voice recognition engine only when voice recognition is necessary, and cancels the voice recognition engine occupancy when voice recognition is not required, such as when transferring a line.
The automatic voice response system according to claim 1. In the automatic voice response system that automatically transfers the phone by specifying the transfer destination by voice response with the computer through the phone,
At the time of response by voice, a flag for determining whether it is recorded voice or synthesized voice is provided, and by managing this flag, a voice that combines recorded voice and synthesized voice can be used,
The automatic voice response system according to claim 1 or 2.