JPH0213166A - Transmitter - Google Patents

Abstract

PURPOSE:To improve the recognition rate and to send an efficient compression signal by sending a code corresponding to a voice of a talker to an opposite center. CONSTITUTION:A voice inputted from a talker by a handset 2 is recognized at a voice recognition section 22 as a word, words and codes correspond one by one to each other in a voice versus code reference table 23, the recognized word string is converted into a code corresponding to the word by using the word string as a retrieval key and sent to a transmitter-receiver 3 via pushbutton public line network 1 from a DTMF generator 24. The transmitter- receiver 3 receives the code from a DTMF receiver 32 and a word corresponding to the code is retrieved by using the code as a retrieval key from the reference table of the same form as the reference table of the voice code. Thus, the operability is obtained, in which a talker can communicate with the transmitter- receiver 3 as if the talker makes conversation with a person regardless that the person is not in existence on the spot.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention is a transmission device that sends a code corresponding to a word string recognized by a speech recognition unit to the other party, and searches the other party for a word corresponding to the code. Regarding.

(Prior Technology and Issues) Conventionally, the first communication device has been used as a communication device for various telephone services that provide so-called seat reservation services and data bank services, etc., in which passengers can reserve seats on airlines, trains, etc. and obtain desired data from data libraries over the phone. A block diagram as shown in FIG. 3 has been proposed.

In FIG. 3, the handset 3 is PI3 (BUTSUSH, I?
The signal pressed by the methane 21 is transmitted from the DTMF' generator and notified to the transmitter/receiver 3 via the PB public line network 1''.The transmitter/receiver 3 is distributed in each region. Each transmitting/receiving device 3 has a handset 2.
A voice recognition/response device 31 is installed to recognize and respond to the sent data. A dial is made to the transmitting/receiving device 3 from the pushbutton/receiving device 21, and when the connection is confirmed, the transmitting/receiving device 3 notifies the caller 4 of a message requesting the operating procedure. For example, the content of the message is XX for button 1, YY for button 2, and ZZ for button 3. The caller 4 must listen to the operating procedure of the button 21 from the transmitting/receiving device 3, record the contents, and then communicate with the transmitting/receiving device 3. With the above configuration, the caller has to listen to and record the operating procedures for each transmitting/receiving device 3, which is cumbersome.

(Problem to be Solved by the Invention 1) The present invention has been made in view of this point, and includes a PB telephone public line network, a voice recognition unit connected to this, and a voice recognition unit that recognizes input voice. a handset and receiver comprising a voice-to-code correspondence table in which a word string recognized by the voice recognition unit and a code corresponding to each word are stored; and a DTMF generator that transmits the code according to the correspondence table; , which is connected to the PB telephone line network and receives the code sent from the transmitting device.
A device (operation) consisting of an 'L' MP receiver, and a transmitting/receiving device that is provided with a correspondence table in the same format as the correspondence table and that searches for the word string inputted by voice corresponding to the code. With the above configuration, the invention includes a handset and a transmitting/receiving device ti.
By providing a voice-to-code correspondence table, the voice input from the transmitting side is recognized by the voice recognition unit as a word string, and the code corresponding to the word string is sent to the other party's center based on the correspondence table. , the partner center searches for the word string that corresponds to that code.
2. The caller should be able to communicate with the transmitter/receiver as if they were interacting with a human being.

(Example) Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a transmission device according to an embodiment of the present invention, and FIG. 2 is a speech-to-code correspondence table diagram explaining the structure of the table used in FIG. 1.

If the name in FIG. 1 is the same as the name explained in the conventional example in FIG. 3, assign the same number.

In FIG. 1, the handset 2 includes a voice recognition unit 22 that recognizes the voice input from a caller 4, and a voice pair code that has a one-to-one correspondence between a word string and a code as shown in FIG. The D 'J' MF generator 24 transmits the code corresponding to the word string to the transmitter/receiver 3, and the transmitter/receiver 3 transmits the code via the PI'3 public line network 1. The DTMF receiver 32 receives the transmitted code, and a table having the same format as the voice-to-code correspondence table described above.

In the handset 1, the voice input from the caller 4 is recognized as a word by the voice recognition unit 22, and as shown in FIG.
) rAl, and includes all tables if there are transmitting/receiving devices 3 from transmitting/receiving device A to transmitting/receiving device Z.

The recognized word string is sent from the DTMF generator 24 which is a code corresponding to the word using the word string as a search key.I) B
The code is transmitted to the transmitting/receiving device 3 via the public public line #iI network 1, and the transmitting/receiving device 3 receives the code from D'1' to the IF receiver 32 without receiving it, and creates a correspondence table as shown in FIG. By using the code as a search key, a word corresponding to the code is searched, and the user 4 can operate the transmitting/receiving device 3 as if he were having a conversation with a human even if there was no human present.

(Effects of the Invention) Since the present invention is configured as explained below,
There are some effective techniques as described below.

Since a code corresponding to the caller's voice is sent to the other party's center, the recognition rate is improved, a shortened and more efficient signal is sent, and a low-cost system can be constructed.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a block diagram of a transmission device in an embodiment of the present invention, and FIG. 2 is a table showing the correspondence between voice and code, which explains the structure of the European transmission table in FIG. 1. FIG. 3 is a block diagram of a conventional transmission device. ... PB public line network 2 ... Handset 1 ... Pushbutton 22 ... Voice recognition section 3 ... Voice-to-code correspondence table 4 ... DTMF generator 3 ... Transmitting/receiving device 1...
Voice recognition response device 2...DTMF receiver

Claims (1)

[Claims] A PB telephone public line network, a voice recognition unit that is connected to this and that recognizes input voice, and a voice pair code that stores word strings recognized by the voice recognition unit and codes corresponding to each word. and a DTMF generator that transmits the code according to the correspondence table; and a DTMF generator that is connected to the PB telephone network and receives the code sent from the communication device. 1. A transmission device comprising: a receiver; and a transmitting/receiving device that includes a correspondence table having the same format as the correspondence table and searches for the word string input by voice corresponding to the code.