JPS61287363A - Method and equipment for voice telephone control - Google Patents

Abstract

PURPOSE:To automate a talking equipment by using a voice processing and voice reception means to confirm a called party name corresponding to a designated code and stored in a storage means and starting a phone call corresponding to a prescribed telephone number. CONSTITUTION:A voice control telephone set 10 consists of an acoustic signal processing section 20, a voice recognition section 40, a control section 50, a display section 60 and a line interface 70. The acoustic signal processing section 20 receives sound information from a microphone 31 of a telephone handset 30. The sound recognition section 40 identified and processes a digital sound signal from the section 20. The control section 50 executes the data transmission/reception between the user and the voice reception section 40. Further, the control section 50 is connected to a standard telephone interface 70 with a display section 60 comprising a LED or the like to execute the software routine stored in a RAM 52, in response to the data flow received by a microprocessor 51. As other circuit, a keyboard 110 is provided to register names of a called party or a talking party by key operation.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a voice telephone control method and device, and relates to the use of operating controls and visual displays in a dialing device and method for establishing a telephone-line connection. It is.

BACKGROUND OF THE INVENTION Many automatic collection telephones are known to provide voice-controlled dialing of telephone numbers.

Problems to be solved by the invention As limitations of conventional automatic collection dialing telephones,
All operations such as registering or changing a new telephone number on the telephone must be done manually, and due to the lack of consideration for the user, it is difficult for beginners as well as those with ordinary operating knowledge to do so. However, it is difficult to use the equipment.

One particular problem is that when users register the names of their interlocutors, each user has a different pronunciation style and sound quality, making it difficult for the telephone to recognize the voice.

Furthermore, one of the problems is that some of the various names to be registered are similar in terms of sounds and letters.
Misrecognition of voices by the telephone and erroneous operations by the user occur.

SUMMARY OF THE INVENTION The present invention provides an inexpensive dialless voice-controlled telephone that uses a computational type visual display to communicate with the user in a simple manner.

The device can recognize many commands such as "Dial,""Bag,""Recall," and "Learn," in which the user tells the device what action he or she wants to perform by voice. These commands appear on the visual display and eliminate noise. To make the telephone extremely easy to use even in many places.

A feature of the invention is that each person with his or her own stored repertoire of names and numbers has the ability to be used by multiple different users, even if there are duplicate names within the repertoire. It is that you are. Additionally, each user can verbally select up to eight names, such as "Tommy," for the device to recognize, and can verbally indicate the corresponding number. You can either speak the number or dial it directly. Either way, users can visually recognize their spoken commands.

Another feature of the invention is that adding and/or changing names and/or numbers from a user's repertoire can be accomplished by voice or keystrokes in response to visually displayed cues. It is something that can be done.

Another feature of the present invention is that the command is a charge code followed by a number, and a "sprint" number followed by a number.
It consists of a series of words and numbers, such as.

This allows the use of alternative communication networks or the creation of computational codes regarding time information.

Furthermore, the most distinctive feature of the present invention is that the name of the called party or interlocutor is registered in the telephone using the keyboard. By registering the name of the called party or interlocutor by key operation, the registration of the name becomes more reliable and the name displayed on the visual display section becomes easier to confirm.

EXAMPLES The present invention will be explained below using examples shown in FIGS. 1 to 4.

As shown in FIG. 1, the voice-controlled telephone device IO according to the embodiment of the present invention includes an acoustic signal processing unit 20. Voice recognition unit 40
．． Control unit 50. Display unit 60 and line interface 7
Consists of 0.

The acoustic signal processing section 20 includes a telephone handset 30 (
Microphone 31 (hereinafter simply referred to as a handset)
A preamplifier (amplifier) 212, a speech equalizer (not shown), a 16-channel analog signal analysis circuit 22, an analog-to-digital converter (A/D converter) (not shown), timing, etc. are adjusted. Consists of peripheral circuits such as buffers.

The speech recognition section 40 is connected to the acoustic signal processing section 20 and performs identification processing on the digital acoustic signal from the acoustic signal processing section 20, and the microprocessor 41. Random access memory (RAM) 42 for storing multiple dedicated user name and called party name data. Erasable and programmable read-only memory (ROM)4
2 and other peripheral circuits.

The speech templates are stored in RAM 42, which can recognize 83 words or phrases supplied by the user.

The microprocessor 41 has a special function to execute firmware stored in the RAM 42, and has a special function of executing the firmware stored in the RAM 42.
Channel analog signal analysis circuit 22 and 8 (not shown)
- Control the bit A/D converter.

Further, the microprocessor 41 can access the voice template stored in the RAM 42 and add functions for changing or deleting the template necessary when adding a new user name, and changing and deleting the name and dial number.

The voice recognition unit 40 is connected to the control unit 5° via a data latch, and the microprocessor 4 of the voice recognition unit 40
1 is input to the control section 5o.

The control unit 50 executes data exchange operations between the user and the voice recognition unit 40.

The control unit 50 includes a general microprocessor 51 and a RAM 52 that stores telephone number data. It is composed of a RAM 54 for storing the name of the called party or the dialogue sound name, and peripheral circuits.

Furthermore, the control unit 50 is connected to a display unit 60 consisting of eight to character light emitting diodes (LEDs) and a standard telephone interface 70, and in response to the flow of data received by the microprocessor 51 and stored in the RAM 52. Execute software routines. RAM5
2 generates information to be displayed and call data to be set or terminated.

The display section 60, as shown in FIG. 2, consists of standard 4-character, 14-segment display elements such as LEDs, and is connected in a conventional manner. microprocessor 5
The data port 1 is connected to a display driver (not shown), and this microprocessor 51 generates a signal to be displayed, as will be described later. Additionally, the microprocessor 51 is connected to the telephone interface 70 via input/output (I10> data latches) and has a standard dialing signal generator (not shown) that is connected to the microprocessor. A dial signal indicating the telephone number of the call to be set up from 51 is received via the latch.

A keyboard 110 is provided as another circuit. This keyboard 110 is connected to a microprocessor 41 and a microprocessor 51 as shown in FIG.

In other words, FIG. 2 corresponds to the device shown in FIG.
a, 22b are audio data preprocessing circuits, 44a, 44
b is a buffer; 43a and 43b are read-only memories (ROM); 42 is a random access memory (RAM) for storing voice data corresponding to the user's name and the name of the called party or interlocutor; and 52 is telephone number data. A RAM 54 stores called party name character data input from the keyboard 110.

Further, 56 is a dialer which sends telephone number data to the telephone interface 7 in accordance with instructions from the microprocessor 51.
Send to 0.

As shown in FIG. 3, the keyboard 110 is provided with keys for inputting numbers O-9 and characters. Note that the keyboard 110 is not limited to the one shown in FIG. 3, and may be any means that can perform both numerical input and character input.

FIG. 4 shows a schematic configuration of a voice-controlled telephone 10 according to an embodiment of the present invention.
Housing 100 containing all the control circuit elements shown in the figure
and 12-keyboard 110. Handset 130. User selection pushbutton switch 160. Redrain pushbutton switch 170 and 8-character LED display 180
It is made up of.

Functions can be performed using pushbutton switches 150, 160, 170 and microprocessor 51.

Operation and Function When the power is turned on, the display section 60 (180) shows the time of 1 minute and 9 seconds under the control of the microprocessor 51.

The six digits of the user number are displayed.

When the handset 120 is picked up to perform an operation and the train button 150 is pressed, a "reset"
mode and the voice control training mode is registered. At the bottom of the “Reso” mode start control, “Dial-4”
" is displayed. Here "4" indicates the number of repetitions,
It signals the user to repeat the word "dial" four times. Every time the word "dial" is uttered into the microphone 130 in the mouthpiece of the handset 120, the voice recognition unit 40 receives the message via the acoustic signal processing unit 20.
will be contacted, and the number of repetition instructions will be from "4" to r3J,
From r3J “2”.

It decreases from "2" to "1" then to "0". In all cases, the user must speak the words displayed until they count down to zero.

The process of displaying the word "dial" and the number of repetitions is controlled by the microprocessor 51 of the control section 50. By controlling this routine, the voice recognition unit 40
The microprocessor 41 is instructed to store it in the AM42. The microprocessor 41 thereby decrements the displayed number by one and causes the speech recognition unit 40 to execute the routine again.

When the word "dial" is recognized four times, RAM4
Among the 17 commands stored in 2, the second priority repeating command number "4" is displayed on the display unit 60 (180). In the train mode, each of 17 command words or 0 to 9 is displayed, and when the user speaks the displayed word or number, the voice recognition unit 40 recognizes the words a total of four times.

One voice-controlled telephone is equipped with a train button 170 that allows the user to train 17 by repeatedly pressing button 170 until a particular word or number appears on display 180 according to the number of repetitions.
A specific one of the command words or numbers can be redrained into the speech recognizer 40.

The user then performs a training process on this particular word or number as in the initial training described above.

Under the control of the ``rL E A RN (Registration)'' routine,
After the user speaks rLEARNJ through the microphone 31, the user further inputs the desired dialogue tone name, such as "TOMy
When (g=-)J is uttered, this audio signal is transmitted to the speech recognition section 40 through the acoustic processing section 20.

In the audio processing unit 20, an amplifier 21 amplifies the analog audio signal, and an audio interface circuit (data preprocessing circuit) 22 converts the analog audio signal into a digital signal and inputs the digital signal to the microprocessor 41 of the audio recognition unit 40. The microprocessor 41 gives a dialogue pitch name display command to the microprocessor 51 of the control section 50. In accordance with this display command, the display section 60 (18G) displays a display rNAME (name) in which the desired dialogue tone name rTOMYJ should be registered. After this display, the user inputs the dialogue tone name rTOMYJ using the keyboard shown in FIG. The procedure for inputting rTOMYJ using the keyboard shown in FIG. 3 is as follows. That is, key "8" has rT U V
J is marked, and to enter "T", press the key "
8" and then press the key "*". key "0" has rO
PRj is marked, and to input "0" manually, press the key "0" and then press the key "*". key "6" is r
M N OJ is marked, and to manually perform rMJ, press the key "6" and then press the key "*". Then the key “
9" is marked with [WXYJ. To input rYJ, press the key "9" and then press the key "#".

After the dialog pitch name rT OM Y J is registered, the display section 6
0 (180) displays an instruction rNo (place) J to register a telephone number.

To register the telephone number, the keyboard 160 may be used or voice input may be used.

After the training process, the display section 6G (180) becomes blank under the control of the microprocessor 51, and the user can key-operate the dial digits one by one. Also, when inputting a phone number by voice, the routine rSE
By controlling TNUMBJ, each digit is communicated to the voice recognition unit 40 and displayed as each digit is recognized as one of ten numbers (0-9) stored in RAM 42. It is visually displayed in section 60. The displayed telephone number is shifted and filled in the display section 60 (180). If a number is displayed incorrectly, the user can correct it by saying rBAcKJ (back) and then saying the correct number.

After the correct telephone number is registered in the voice telephone 10, rR
Under the control of a routine starting at location EcALLJ,
When the user wants to call his/her conversation partner "Tommy", the user must say rTOMYJ. rTO
If MYJ is pronounced correctly, rTOMYj will be displayed on the display section 60 (180). After confirming this display, the user calls the corresponding telephone number. When calling a telephone number, the telephone number is displayed on the display section 60 (18
0), say the words rRE CAL L J. Furthermore, after the word RECALLJ, if the user says ``DIAL'' four times in a row, a telephone call will begin. That is, the control unit uses the telephone interface 7
0 for a predetermined period of time, e.g.
The automatic call will begin in seconds.

Since the interlocutor's name rTOMYJ is registered by key operation, rT ON may occur due to inaccuracies in the user's voice or pronunciation.
Erroneous registration of Y, J, etc. is prevented. Furthermore, since the name of the called party is first displayed when dialing, the user can perform a predetermined dialing operation after confirming this display, thereby preventing erroneous dialing. The dialing operation is performed automatically, but for greater accuracy it may also be by keyboard operation or by voice recognition means.

DETAILED DESCRIPTION OF THE INVENTION According to the present invention, when the called party data is accessed by voice, the called party's name is displayed on the display. You can prevent wrong calls.

Furthermore, the most outstanding advantage of the present invention is that since the user's or interlocutor's name is registered by key operation, it is possible to prevent incorrect registration of the telephone itself due to unclear voice, and to improve reliability. I can do it.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a voice-controlled telephone device according to the present invention;
FIG. 2 is an electrical wiring diagram showing a part of the device shown in FIG. 1, FIG. 3 is a schematic configuration diagram showing an example of a keyboard used in the voice-controlled telephone device of the present invention, and FIG. 4 is a telephone to which the present invention is applied. FIG. 20... Sound processing unit, 21... Preamplifier, 22...
...Voice data preprocessing circuit, 30...Handset,
40... Voice recognition unit, 41... Microprocessor, 50... Control unit, 51... Microprocessor, 60
(180) Display unit, 70 Telephone interface, 1
10...Keyboard.

Claims (5)

[Claims] (1) Switch means for manually selecting either a normal operating mode or a programming mode for operation; a digital command signal indicating one of a predetermined set of operating conditions; and the name or telephone number of the person registered. voice signal processing and recognition means responsive to speaking an input word for generating a digital signal indicative of a number; and voice signal processing and recognition means responsive to said command and data signals and for storing and recognizing a telephone number from a common carrier phone system. said mode for generating standard signals and causing a visual display of said operating status, said command data signals and information for initiating and completing calls associated with a telephone subscriber at a particular telephone number; a digital processor and memory means responsive to the switch means; keyboard means for storing in said memory means input data of at least the name of the called party; and said operating status, command data signal, name and telephone number of the telephone subscriber. visual display means for displaying information to a user of the dialing device in response to a visual display signal generated from said microprocessor; 1. A voice telephone control device comprising means for confirming the name of a called party and initiating a telephone call in response to a predetermined telephone number. (2) In a voice-controlled telephone device that can be operated in either a normal mode or a program creation mode for operation, which includes a voice recognition means for recognizing a registered predetermined input word, the telephone line connection is performed by the following steps. A voice telephone control method characterized in that: a) Manually selecting the programming operation mode. b) displaying on a visual display means a predetermined appropriate word input by a predetermined interlocutor; c) Pronounce the appropriate word a number of times equal to the repetition number after each person has recognized the pronunciation of the appropriate word until the number reaches 0 and decrease the number; d) Repeat steps (b) and (c) a predetermined number of times. e) Register normal mode operations. f) Entering the name of the telephone subscriber by keyboard means. The input and recognized name of the registered person is displayed on the visual display section. g) recognizing the displayed registered name of the person to be dialed and causing a telephone connection to take place by calling the name displayed on the visual display; h) Speak the name corresponding to the predetermined designation code set in step (f) above. i) Make a call operation corresponding to the name recognized in step (h). (3) The voice telephone control method according to claim 2, wherein the calling operation is performed using the keyboard means in correspondence with a predetermined registered name. (4) The voice telephone control method according to claim 2, wherein the telephone call operation is performed by the voice recognition means in accordance with a predetermined registered name. (5) The voice telephone control method according to claim 2, wherein the calling operation is automatically performed in accordance with a predetermined registered name.