JPH0332148A - Voice recognition telephone set - Google Patents

Abstract

PURPOSE:To recognize a voice signal on a telephone line and to store only the required part to a storage device by providing a recognition device recognizing the voice signal, a recognition button and a storage device. CONSTITUTION:A telephone set circuit 11 and a voice signal recognition device 12 are connected to a telephone line 15 and a signal on the telephone line 15 is converted into a digital signal by a signal conversion section 16 of the voice signal recognition device 12 and converted into a data in matching with a form of a voice recognition dictionary 17 and recognized with a recognition control section 18 based on a data in the voice recognition dictionary 17. While a recognition button 13 is depressed, the result of recognition is written from a recognition gate 19 to a storage device 14. Thus, the voice recognition dictionary 17 limited to a specific data coverage such as addresses is prepared and while an opposite party utters an address, when the recognition button 13 is depressed, the recognition with high recognition precision is attained and only a desired data is stored in the storage device 14.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a telephone having a function of recognizing voice signals.

``Prior Art'' Conventionally, the primary function of telephones has been the ability to transmit voice signals. Since voice signals are temporary, it is necessary for the user to interpret the information heard from the other party and record it in a memo. For example, information such as the other party's address must be confirmed including kanji characters, which is cumbersome. One way to solve this problem is to use FAX (Facsimile 9) to receive text information directly from the other party, but in this case, both parties must use a FAX terminal or a terminal that can receive FAX signals. It is necessary to have Furthermore, when asking for the address, whether by telephone or fax, it is necessary to transcribe it again, which is cumbersome.

Recently, telephones with a voice recognition function have been released, but they do not recognize the contents of a single call, only by making a call to a pre-registered party or executing a specific command by voice.

An object of the present invention is to provide a telephone with a voice signal recognition function that overcomes the transient nature of voice signals.

"Means for Solving the Problem" According to the present invention, a voice signal on a telephone line is recognized by a recognition device, and the recognition result of the recognition device is stored in a storage device only while a recognition button is being controlled.

Embodiment FIG. 1 is a diagram for explaining a first embodiment of the present invention, in which a telephone circuit 11, a voice signal recognition device 12, a recognition button 13, and a recognition result of the voice signal recognition device 12 are shown. It consists of a storage device 14 for storing data.

The telephone circuit 11 and the voice signal recognition device 12 are connected to the telephone line 1
5 (= connected.

As shown in FIG. 2, the internal configuration of the speech signal recognition device 12 includes a signal conversion section 16, a speech recognition dictionary 17, and a recognition control section 1.
8 and a recognition game 19 for storing the recognition results in the storage device 14.

In the 22 that operates this, the signal on the telephone line 15 is converted into a digital signal by the signal converter 16ζ2.

Furthermore, the format ζ of the two speech recognition dictionaries 17 is converted into combined data, and this converted data is recognized based on the data of the speech recognition dictionary 17 by the function C2 of the recognition control unit 18. while being pressed.

The recognition result is written from the recognition gate 19 to the storage device 14 .

Since it has such a structure, the recognition button 131 =
Therefore, during the designated time, signals on the telephone line can be recognized based on the voice recognition dictionary 17 and stored in the recording device 14.

The effect is that by preparing a voice recognition dictionary 17 limited to a specific range, such as an address, and pressing the recognition button 13 while the other party is speaking the address, recognition with high recognition accuracy becomes possible. And only the desired items are stored in the storage device 14.
can be memorized.

FIG. 3 is a diagram showing a second embodiment of the invention, and parts corresponding to those in FIG. 1 are given the same reference numerals. In this example, a printing device 21 for printing the contents stored in the storage device 14 is provided.

In order to operate this, the telephone line 1 must be recognized while the recognition button 13 is pressed and stored in the storage device 14.
5 is read out from the storage device 14 and printed by the printing device 21.

Because of this effect, the audio signal on the telephone line 15 is printed.

One advantage of this is that, for example, the address information uttered by the person on the other end of the call can be directly printed, saving the effort of transcribing it from a memo. Furthermore, if the speech recognition dictionary 17 is created specifically for an address, it becomes possible to easily write an address even if the address uses special kanji or has a special reading.

FIG. 4 is a diagram showing a third embodiment of the present invention, in which a confirmation device 22 for confirming the recognized result is added to the configuration of FIG. 1. As shown in FIG. 5, the confirmation device 22 is equipped with a display device 23 and an operation button 24, and each of the cursor keys marked with "(〃)" in the operation button 24 is a cursor on the display surface of the display device 23. 25, the conversion key with ``conversion'' is used to display the next candidate, and the confirmation key with ``confirmation'' is used to indicate the final confirmation.

The internal configuration of the confirmation device 22 is shown in Figure WJ6. A confirmation device control section 26 is connected to the display device 23 and the operation buttons 24, and is further connected to the storage device 14.

To operate this, the recognition results stored in the storage device 14 along with the most likely recognition result candidates are displayed on the display R23 (2).Use the cursor key # (or ")1 to move the cursor 25 to the incorrectly recognized character. Move and use the ``Convert'' key to display the next candidate.

Correct the incorrectly converted characters one by one, and when the correct character is finally obtained, press the Enter key to confirm. These functions are executed by the functions of the confirmation device control section 26. Because it works like this, recognition results can be checked and corrected one after another. The effect is that even if an audio signal is misrecognized, it can be corrected.

FIG. 7 shows a fourth embodiment of the present invention, in which an external system adapter (R8232C interface) 27 is added to the configuration of FIG.

In order to operate this, the information stored in the storage device 14 is connected to an external personal computer (personal computer) via an external system interface (tR8232c in this example), and the computer control C: Therefore, the storage device 14
Communicate information stored in. Because of this effect, the information stored in the storage device 14 can be provided to an external personal computer.

As a result, it becomes possible to use the information obtained during a telephone call in an address book management system, etc., which manages the information on an external computer.

Although the recognition gate 19 is opened only while the recognition button 13 is pressed in the above description, the recognition gate 19 may be omitted and the audio signal recognition device 12 may be operated only while the recognition button 13 is pressed.

"Effects of the Invention" As explained above, according to the present invention, the voice signal on the telephone line is recognized and only the necessary part is stored in the storage device. It has the advantage of being able to be edited, printed and addressed from the contents of the storage device, and transferred to an external system for use with external application software.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a first embodiment of the present invention, FIG. 2 is a block diagram showing the internal configuration of the audio signal recognition device 12, and FIG. 3 is a block diagram showing the configuration of the second embodiment of the invention. FIG. 4 is a block diagram showing the configuration of a third embodiment of the present invention. FIG. 5 is a block diagram showing an example of the external configuration of the confirmation device 22;
FIG. 6 is a block diagram showing the internal configuration of the confirmation device 22, and FIG. 7 is a block diagram showing the configuration of a fourth embodiment of the present invention.

Claims (1)

[Claims] (1) A telephone comprising a recognition device that recognizes voice signals on a telephone line, a recognition button, and a storage device that stores the recognition results, wherein the recognition device is activated only while the recognition button is controlled. A voice recognition telephone, characterized in that recognition results are stored in the storage device.