JP2001308970A - Speech recognition operation method and system for portable telephone - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a speech recognition operation method for portable telephone and its system by which incoming speech and outgoing speech are possible by dialing the portable telephone by a speech instruction word. SOLUTION: A speech recognition technology is utilized to allow the portable telephone to recognize simple instruction words and entered keys such as a telephone number for the operation of the portable telephone in speech and to enter the telephone number, the portable telephone has a function of executing incoming and outgoing calls and interruption of the incoming call and the outgoing call and deletes a telephone number that is mis-recognized or the like by a speech instruction word, the presence of operation of a speech recognition processor is selectively decided to reduce power consumption depending on the type of the consumed power supply in use. Adding the voice processor and in interface circuit provided with a simple logical arithmetic operation to a conventional portable telephone system can attain control in voice without circuit revision, so as to enhance the convenience of operation during driving or the like and this system is applicable to a telephone set for a general home or a public telephone.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a mobile phone,
In particular, the present invention relates to a voice recognition operation method and a system for a mobile phone capable of dialing by a voice command and performing an incoming call and an outgoing call.

[0002]

2. Description of the Related Art In the case of general cellular phones and car phones, when trying to make a call from a state where the power is on, a cover switch of the cellular phone is opened or a handset of the car phone is lifted. And put it in the call waiting state,
By pressing all the telephone numbers of the parties to be called and pressing the SEND button, the call is dialed and the call can be made. Also, when trying to answer an incoming call in the reception standby state, a call can be made by opening the cover switch or pressing the send button. If a wrong number is entered by entering the number to be called, the incorrectly entered telephone number must be deleted by pressing the CLEAR button.

The operation method of opening the cover switch using the hand and the eye or pressing the button while checking the input button one by one is inconvenient for the operation while the user is driving. Similarly, there is considerable inconvenience in using a mobile phone when performing other tasks with hands or when a user is blind or has limited mobility.

[0004]

SUMMARY OF THE INVENTION Accordingly, the present invention has been devised to solve the above-described problems, and is directed to a simple command for operating a mobile phone using voice recognition technology. Input keys such as words and numbers of phone numbers are recognized by voice, and the phone numbers are dialed and incoming and outgoing calls can be disconnected and incoming and outgoing calls can be disconnected. Register the relevant department names distinguished by each other or multiple extension numbers, each regional name and country name distinguished by multiple regional numbers by voice, and the registered name corresponds to the corresponding telephone number, extension number and region A voice recognition operation of a mobile phone that can perform a speed dial function by voice recognition of letters and numbers by registering numbers and country codes (including international telephone connection numbers for each communication carrier) As its object to provide a legal and system.

Another object of the present invention is to delete a wrongly or error-recognized telephone number by a voice command, and selectively determine whether a voice recognition processor is activated according to the type of power consumption used. Another object of the present invention is to provide a method and a system for voice recognition operation of a mobile phone, which can reduce power consumption.

[0006]

According to a first aspect of the present invention, there is provided a voice recognition operation method for a portable telephone, comprising: (a) voice recognition of a power-on command in a reception standby state; (B) in a call waiting state, (b) in a call waiting state, (b) in a call waiting state, A step of dialing the telephone number and converting to a busy state when the telephone number and the send command continuously input within a predetermined time are recognized; and (c) a call stop command in the busy state. , And ends the call and switches to a reception standby state.

According to a second aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect.
If the stop clear command is recognized by speech at the stage,
The method may further include a step of returning to the call waiting state.

According to a third aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect of the present invention.
The steps are as follows: the entire telephone number is recognized and stored, and the stored telephone number is reproduced and output. If the stop clear command is recognized after the reproduction and output, an error occurs in the input telephone number. Is determined to have occurred, the entire stored phone number is deleted and the call is switched to the call waiting state again.
Dialing the stored telephone number.

According to a fourth aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect of the present invention.
And controlling the on / off of the speech recognition processor according to the output signal of the on / off switch of the speech dialing when the power consumption of the speech recognition processor performing the step (c) is used by the internal power supply. It is a feature.

According to a fifth aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect of the present invention.
When the power consumption of the speech recognition processor performing step (c) is used as an external power supply of the hands-free socket, the speech recognition processor is continuously driven by the external power supply. is there.

According to a sixth aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect of the present invention.
The reference voice pattern stored in the voice recognition processor performing the step (c) is registered by receiving the voice command and the number and symbol of the telephone number in the voice of the user. Things.

According to a seventh aspect of the present invention, there is provided a voice recognition operation method for a portable telephone according to the first aspect of the present invention.
The reference voice pattern stored in the voice recognition processor performing the steps (c) is a voice command, a number and a symbol obtained by an average pronunciation of many people.

[0013] According to a second aspect of the present invention, in addition to the configuration of the first aspect, the voice recognition operation method for a portable telephone according to the eighth aspect of the present invention further comprises:
In the call waiting state, when a character that is continuously input together with a specific character specified in advance for the speed dial function and a send command are speech-recognized within a predetermined time, the corresponding step corresponds to the recognized character. The method further comprises the steps of: searching for a telephone number to be called, automatically shortening the searched telephone number, and switching to a busy state.

According to a ninth aspect of the present invention, in addition to the configuration of the eighth aspect, the characters recognizable in the call waiting state are subscriber name, trade name, domestic and domestic names. It is characterized by including the name of the region outside the country and the signature of the related department.

According to a tenth aspect of the present invention, in addition to the configuration of the ninth aspect, the abbreviated dialing step in the outgoing call waiting state includes an international call classified by a communication carrier. The dialing includes a connection number, a country code classified by each country, a domestic and foreign area code, and a telephone number of a subscriber.

The voice recognition system for a portable telephone according to the present invention is characterized in that an input pattern obtained by extracting a characteristic parameter from a voice signal including an input voice command and a telephone number and a preset reference. A voice recognition processor that measures the similarity to the pattern and recognizes the input voice; and the same as the signal value recognized by the voice recognition processor while being connected to the voice recognition processor, the handset key input unit, and the cover switch. An interface unit that logically combines the output signal value of the handset key input unit and the output signal value of the cover switch having a function, that is, the voice recognition processor, the handset key input unit, the cover switch, The output signal value of the handset key input section is An interface unit logically associated with the corresponding signal value recognized by the server, storage and deletion of the telephone number recognized according to the combination signal provided from the interface unit, dialing of the telephone number, start of the call and A control unit for controlling the general operation of the mobile phone related to the end of the call.

According to a twelfth aspect of the present invention, in addition to the configuration of the eleventh aspect, the voice recognition processor selectively supplies the internal battery power by turning on / off a selection switch. , And driven.

According to a thirteenth aspect of the present invention, in the voice recognition system for a portable telephone according to the eleventh aspect, the voice recognition processor is supplied with power from an external power source input through a hands-free socket. It is characterized by being driven.

According to a fourteenth aspect of the present invention, in the speech recognition system for a portable telephone according to the eleventh aspect, the speech recognition processor converts the voice command and the telephone number by a user's personal voice. It is characterized by comprising a memory storing a reference pattern generated upon receiving an input.

According to a fifteenth aspect of the present invention, in the voice recognition system for a portable telephone according to the eleventh aspect, the voice recognition processor acquires the voice command obtained from an average pronunciation of many people. And a memory for storing a reference pattern generated by a telephone number. In other words, in this speech recognition system,
An average pronunciation is analyzed based on a voice command and a voice signal of a telephone number uttered in advance by a plurality of persons, and a reference pattern is obtained from the average pronunciation.

According to a sixteenth aspect of the present invention, in the mobile phone voice recognition system according to the eleventh aspect, the voice recognition processor includes a voice command, a pre-registered character and a telephone number. The method is characterized in that a similarity between an input pattern obtained by extracting a characteristic parameter from a voice signal and a preset reference pattern is measured to recognize a voice.

According to a seventeenth aspect of the present invention, in addition to the configuration of the sixteenth aspect, the voice recognition processor includes means for detecting a connection state of an external power source / internal battery power source. It operates only when the external power supply is connected. That is, in this speech recognition system, the speech recognition processor is driven only when the external power supply is connected.

The speech recognition system for a portable telephone according to claim 18 of the present invention, in addition to the constitution of claim 16, wherein the speech recognition processor is installed outside the telephone through an interface means extended by a hands-free socket. Connected to the control unit and driven by receiving power supply from an external power supply.

A speech recognition system for a portable telephone according to the present invention, comprising: a hands-free socket having a speech recognition processor for recognizing a speech command and a telephone number inputted via a microphone; The free socket is wired / wirelessly connected and communicates with each other, and receives a voice command and a telephone number recognized by the voice recognition processor to receive a telephone number dialing, a call start, and a call end according to the command. And a handset for performing the relevant operation.

According to a twentieth aspect of the present invention, in the mobile phone voice recognition system according to the nineteenth aspect, the voice recognition processor may further include a user's personal voice for converting the voice command and the telephone number. It is characterized by comprising a memory for storing a reference pattern generated in response to an input.

According to a twenty-first aspect of the present invention, in the speech recognition system for a portable telephone according to the nineteenth aspect, the speech recognition processor acquires the speech command obtained from an average pronunciation of many people. And a memory for storing a reference pattern generated by the telephone number and the telephone number.

According to a twenty-second aspect of the present invention, in the speech recognition system for a portable telephone according to the nineteenth aspect, the speech recognition processor associates a subscriber name and a mutual name with a domestic and foreign region name. The department name is registered in advance, and the pre-registered characters are international telephone connection numbers classified by telecommunications carrier, country codes classified by country, domestic and foreign area codes, and subscriber telephone numbers. Are stored in correspondence with each other.

[0028]

Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

The conventional mobile phone is roughly divided into a transmitting section, a receiving section, a logic section, an antenna section, a display section, and a power supply section. According to the present invention, a voice recognition processor is additionally provided in the conventional mobile phone system. That is, various operations are operated by voice recognition without changing the conventional system.

FIG. 1 is a diagram showing a system configuration of a portable telephone according to the present invention.

The control unit 100 includes a reception signal processing unit 110,
It is connected to the transmission signal processing unit 120, the power supply unit 130, the memory unit 140, and the interface unit 180,
Controls the overall operation of the mobile phone.

The reception signal processing section 110 is controlled by the control section 100, and receives and demodulates a radio signal from an antenna (not shown). The demodulated signal is output as a voice signal through the receiver 111. The transmission signal processing unit 120 is controlled by the control unit 100, modulates an audio signal input from the transmitter 121 into a radio signal, and transmits the radio signal via an antenna.

The power supply unit 130 can be supplied from an internal battery power supply of the handset or an external power supply of the hands-free socket, and supplies power to each unit under the control of the control unit 100. The memory unit 140 is configured as a nonvolatile / volatile memory that stores a main processing program of the control unit 100 and temporarily stores various data generated by processing the main processing program.

The interface unit 180 is connected to the voice recognition processor 150, the handset key input unit 160 and the cover switch 170, and logically combines the output signals of the voice recognition processor 150, the handset key input unit 160 and the cover switch 170, It is provided to the control unit 100 in association with it. The logical correspondence is obtained by performing an OR operation or an AND operation on the output signal value of the handset key input unit or the output signal value of the cover switch having the same function as the signal value recognized by the speech recognition processor. By providing the control signal to the control unit, the control unit receiving the output signal performs the same operation when the button is pressed or the command word or the number is pronounced. Thus, the interface unit 1
80 manages the operation of the handset key input unit 160 and the cover switch 170 in association with the signal value recognized by the voice recognition processor, and provides the signal value to the control unit in accordance with these operations, thereby providing voice control. The same function can be performed in either case by a signal or by a button operation or cover switch operation.

The speech recognition processor 150 includes the transmitter 1
The speech signal (eg, voice command (power on, stop clear, send, call, call stop,
Recognize abbreviated keys), pre-registered characters, and voice handset key values (numbers, symbols).

The speech recognition processor 150 can actually be embodied as software or as hardware. When embodied in hardware, the illustration is omitted, but the speech recognition processor 150 uses the reference pattern (speech command, registered character,
A first memory storing a handset key value);
Various parameters of the voice input through the transmitter 121 are extracted to create an input pattern, and the similarity between the input pattern and a preset reference pattern is measured (a type of pattern matching) to determine the input voice. A processing unit for recognizing, a second memory for temporarily storing the voice recognized by the processing unit, a telephone number and a local number of the subscriber corresponding to the registered characters and each character, and an extension number and communication used for a key phone A third memory is provided for storing connection numbers and country codes of international calls for each worker.

On the other hand, in a preferred embodiment, the reference pattern can be constructed by a speaker dependent system (specific speaker system) and a speaker independent system (unspecified speaker system). In the case of the speaker-dependent system, the voice command words, pre-registered characters, phone number, sharp (#), star (*), etc. are pronounced and directly registered by the user's own voice. EE personal voice reference pattern
When the mobile phone is used by storing it in a memory similar to the PROM and using the mobile phone, the accuracy of voice recognition is more than 90%, which is effective. In the case of the speaker-independent method, voice commands collected from many people, registered characters and digits of telephone numbers,
A mobile phone can be manufactured by storing an average pattern such as a sharp (#) or a star (*) in a memory in advance.

As described above, the power supply of the speech recognition processor 150 is performed by controlling the power of the power supply
May be provided through the interface 180 under the control of the user.

On the other hand, in another embodiment, the speech recognition processor 150 may be driven directly by receiving an internal battery power or an external power. The voice recognition processor may include means for detecting a connection state of an external power supply or an internal battery power supply, respectively, so that the voice recognition processor operates only when the external power supply is connected and uses the internal battery power supply. In this case, a selection switch for voice dialing is mounted to connect or disconnect the internal battery power of the handset. This has the effect of minimizing the power consumption of the speech recognition processor due to the trend toward smaller and lighter mobile phones and longer battery life.

When an external power source such as a car battery is used, power is supplied to the voice recognition processor to drive it unconditionally while the handset is inserted into the hands-free socket. Thus, while the external power is supplied, the power is continuously supplied to the speech recognition processor, and the driving state is maintained. This means that not only is the external power source sufficient,
A voice recognition dial function is required, so if an external power supply is used, it will be driven immediately without turning on the voice recognition selection switch, and if the external power supply is removed, it will be driven according to the on / off control of the voice recognition switch By doing so, the convenience of the user can be enhanced.

First, an embodiment of a method of operating a mobile phone by voice recognition will be described with reference to FIGS. FIG.
FIG. 3 is a state diagram according to the method of voice recognition operation of the mobile phone of the present invention, and FIG. 3 is a flowchart diagram of the method of voice recognition operation of the mobile phone of the present invention.

FIG. 2 shows a process in which one of the five states that can be generated when the mobile phone is operated is changed to another state by a user's voice.
The operation indicated by the dotted line is the operation performed by the conventional button input operation, and the operation indicated by the solid line is the operation performed by the voice input of the present invention. The words used in speech recognition used here cannot be words in the content of ordinary dialogue, and must be used in terms of sampling for speech recognition. We need to know that the term must be memorable.

The description for each state is as follows.

(1) If the cover switch of the mobile phone is opened in the reception standby state, the operation state of the mobile phone becomes the transmission standby state. As described above, even if the cover switch is not opened, if the user speaks the power command "power on" by voice, the state is switched to the call waiting state. The term "power on" refers to the power of turning on a mobile phone (abbreviation, PW
R) is different from the term. That is, "power on"
The command is a preparation command for making a call from a state where the power of the mobile phone is turned on.

(2) In a call waiting state, by pressing a button, a telephone call can be made by pressing a desired telephone number and then pressing a <SEND> button. Instead, if you want to make a voice call, enter the desired phone number within a certain period of time, "1" for "ichi", "2" for "d", "3" for "sun", "4". Is “Yon”, “5” is “Go”, “6” is “Roku”, “7” is “Nana”, “8” is “Hachi”,
Speak “9” using the terms “Kew”, “0” using “Rey”, “*” using “Star”, and “#” using the terms “Sharp” before speaking the “SEND” command. Switch to playback mode. At this time, in the reproduction mode, the recognized and stored telephone number is reproduced and output again to be approved by the user. If the voice recognition processor correctly recognizes the input voice and speaks the "send" command, the recognized telephone number is dialed and the call is switched to a call state.
If the speech recognition processor fails to recognize the speech and speaks the "stop clear" command, the speech recognition processor returns to the call waiting state.

(3) In the reception standby state, if a bell normally rings, a call can be made by opening the cover of the mobile phone.
If the bell rings when the mobile phone cover is open, you can talk by pressing the <SEND> button. Alternatively, in order to switch to the call state by voice, the bell is switched to the ringing state if the bell ringing is detected in the reception standby state. At this time, the state is switched to the talking state if "call" is instructed. Become like That is, it is possible to switch to the call state only when the command "call" is issued while the bell rings. This has nothing to do with opening and closing the cell phone cover.

(4) To end the call with the other party in the call state, the user must press the <END> button. Alternatively, if the user instructs to "stop the call" by voice, the call is terminated and the apparatus returns to the reception standby state.

(5) In a call waiting state, if the telephone number is normally pressed incorrectly, if the <CLEAR> button is pressed once briefly,
Only some numbers are cleared, and if you hold down a bit longer, the entire number is cleared. Instead, during speech recognition,
If you misspell your phone number or your voice is recognized as an error,
If this is deleted and a call must be made again, a command "STOP CLEAR" is spoken, and the system returns to the call waiting state and recognizes the next telephone number.

Referring to FIG. 3, processing of the mobile phone according to the embodiment of the present invention will be described. Here, it is assumed that the voice recognition switch is turned on when the internal battery power is used, or that the voice recognition processor 150 is operated using the external power.

First, in the reception standby state of step 300,
The control unit 100 shown in FIG. 1 detects the incoming call control signal input through the reception signal processing unit 110 and outputs a bell sound signal. At this time, in step 311, the control unit 100 checks whether the voice recognition processor 150 recognizes the voice input through the transmitter 121 as a “call” command.
Here, the recognition result signal of the voice recognition processor 150 is transmitted to the control unit 100 through the interface unit 180. The interface unit 180 performs a logical product operation of the bell ringing detection signal and the voice recognition pulse value of the "call" command. And provides it to the control unit 100.

If it is not recognized as a "call" command, a corresponding process is performed by the control program (320),
If recognized by the "call" command, a known call service is provided (312).

In step 313, the control unit 100 checks whether the voice recognition processor 150 recognizes the input voice as a "call stop" command. If the command is recognized as the "call stop" command, the call in progress is terminated in step 314, and the process proceeds to a reception waiting state.

On the other hand, when the ringing tone of the incoming call ring is not detected while the apparatus is in the reception standby state in steps 300 and 310, the control unit 100 controls the speech recognition processor 150 to output the speech inputted by the speech recognition processor 150 in step 330. on"
Check whether it is recognized as a command. Here, the recognition result signal of the voice recognition processor 150 is transmitted to the control unit 100 through the interface unit 180. The interface unit 180 logically combines the output pulse value of the cover switch 170 and the voice recognition pulse value of the "power-on" command. The sum is calculated and provided to the control unit 100.

If the command is not recognized as a "power on" command, the corresponding operation is performed according to the control program in step 320. If it is recognized by the “power-on” command, the control unit 100 proceeds to step 331 to
A check is made to see if the 50 input voices have been recognized as telephone numbers, Arabic numerals. If it is not a number, step 331 is further performed, and if it is a number, the telephone number whose speech has been recognized is stored in a memory.

On the other hand, during the progress of step 332, the control unit 100 checks whether the voice input from the voice recognition processor 150 has been recognized as a "stop clear" command (333).

If it is recognized as a "stop clear" command, a series of telephone numbers are pronounced one by one,
The user pronounces the number while storing them one by one, judges that the number is wrong, deletes the entire number of the stored telephone number, and proceeds to step 331 again, and starts from the first number. Get input again.

If it is not recognized as the "stop clear" command, the control unit 100 checks whether the voice recognition processor 150 recognizes the input voice as the "send" command (334). If the command is not recognized as a "send" command, the process proceeds to step 320 to perform a corresponding process. On the other hand, if the command is recognized as the "send" command, in step 335, the control unit 100 reads the telephone number stored in the memory, reproduces the telephone number, and outputs it.

In still another embodiment, for example, if a call is made to the 7000 number of the area code 02, 700 station in Seoul, "ray, ni, nana, ray, ray, nana, ray, ray, ray, send "", "Nana, Nana, Ray, Ray, Nana, Ray, Ray, Ray, Send" Must be removed to recognize only the numbers.

In step 336, the control unit 100 determines whether the input speech of the speech recognition processor 150 is recognized as a "send" command. If it is not the "send" command, the voice recognition processor 150 recognizes the number of the telephone number, judges that it is wrong, and returns to step 331 to receive the input again. If the command is recognized as the "send" command, the recognized telephone number is determined to be correct in step 337, and the approved telephone number is dialed. Here, the recognition result signal of the speech recognition processor 150 is transmitted through the interface unit 180 to the control unit 10.
0, the interface unit 180 performs a logical OR operation on the output pulse value of the <SEND> button of the handset key input unit 160 and the speech recognition pulse value of the “Send” command, and sends the result to the control unit 100. By providing
The same operation is performed in both cases.

Then, while exchanging signals between the radio base station and the called terminal according to the normal call setup protocol, a channel is allocated and a communication path is established. At this time, in step 338, the control unit 100 determines whether the call connection is successful, and if not successful, performs step 320.

In step 338, if the call connection is successful, the control unit 100 proceeds to step 339 to perform a call service. During the call service, the control unit 100 switches the voice input from the speech recognition processor 150 to “call stop”.
Check whether it is recognized as a command. If the control unit 100 recognizes that the command is a "call stop" command, the control unit 100 proceeds to step 34.
In step 1, the current call is terminated, the normal call release procedure with the radio base station is performed, and the process enters the reception standby state in step 300 again.

In another embodiment, a hands-free socket may be embodied by incorporating a speech recognition processor for performing a speech recognition function.

If a voice command or a telephone number inputted through the microphone of the hands-free socket is recognized through the built-in voice recognition processor, the recognized voice command and the signal value of the telephone number are It is provided to the control unit of the handset in a wired (wireless) manner through a connector (interface) between the handsfree socket and the handset. The configuration and operation of the voice recognition processor built in the hands-free socket and the operation of the mobile phone by the control unit of the handset are described with reference to FIGS. 1, 2, and 3 above. And its detailed description is omitted.

FIGS. 4 and 5 are flow charts for explaining another embodiment of the method of operating a portable telephone by voice recognition. The following operations are performed when the internal battery power is used, the voice recognition switch is turned on, or the voice recognition processor 150 is connected to an external power source such as a car.
The description will be made assuming that is operated.

First, referring to FIG.
If the incoming call bell signal is detected in the reception standby state of
In step 411, the control unit 100 checks whether the speech recognition processor 150 recognizes the speech input through the transmitter 121 as a "call" command.

If the command "call" is not recognized, the control program performs a corresponding process set in advance (440). If the command is recognized as "call", the incoming call is connected in a known manner. (412). At this time, if the call connection is successful (branch to “Yes” in 413),
The process proceeds to step 413 to provide a call service for an incoming call (431, 432). Next, the control unit 10
0 checks whether the voice recognition processor 150 recognizes the voice input through the transmitter 121 as a "call stop" command (433). If "call stop"
If it is recognized as a command word (branch to "Yes" in 433), the call in progress is terminated in step 434, and the process returns to the reception standby state.

On the other hand, in steps 400 and 410, when the receiving unit is in the reception standby state and the ringing sound of the ringing bell is not detected, in step 420, the control unit 100 switches the voice inputted by the voice recognition processor 150 to " Power on "
Check whether it is recognized as a command.

If it is not recognized as a "power-on" command, a corresponding process is executed in step 440 according to the control program. If the control unit 100 recognizes the power-on command, in step 421, the control unit 100 checks whether the voice input to the voice recognition processor 150 has been recognized as Arabic numerals which are telephone numbers. If it is not a number, the process proceeds to FIG. 5 to perform a speed dial function using a pre-registered character, and then performs a processing step 450 for recognizing a specific character ("abbreviated key") command,
If it is a numeral, the telephone number recognized by voice is stored in the memory (422).

On the other hand, during the progress of step 422, the control unit 100 checks whether or not the speech recognition processor 150 has recognized the inputted speech as a "clear" command (423).

If it is recognized as a "clear" command, a series of telephone numbers are pronounced one by one, and the user pronounces the numbers while storing these one by one, and a mistake is made. Then, the whole number of the currently input telephone number is deleted (424), and the process proceeds to step 421 again so that the number of the telephone number can be received again.

If the command is not recognized as a "clear" command, the control unit 100 checks whether the voice input to the speech recognition processor 150 is recognized as a "send" command (425). If the command is not recognized as the "send" command, the process proceeds to step 421 to perform an operation of determining whether the numeral or the pre-registered character of the telephone number is recognized again. On the other hand, if it is recognized as the "send" command, in step 426, the control unit 100 reads the telephone number stored in the memory, reproduces the telephone number by voice, and outputs it.

In step 427, the control unit 100 checks whether the voice input to the voice recognition processor 150 has been recognized as a "send" command. If "Send"
If it is not a command word, the speech recognition processor 150 recognizes the number of the telephone number and judges that it is wrong, and step 421
Proceed to. If it is recognized as "Send" command, step 4
At 28, it is determined that the recognized telephone number is correct, and the approved telephone number is dialed. Then, while exchanging signals between the wireless base station and the terminating terminal according to the normal call setup protocol, a channel is allocated and a communication path is established (429). At this time, the control unit 100 determines whether the call connection is successful in step 430, and if not successful, returns to step 427 for retransmission and proceeds from step 428 to step 430.
Repeat until.

At step 430, if the call connection is successful, the control unit 100 proceeds to step 431 to switch to a busy state and perform a call service (432). During the call service, the control unit 100 checks whether the speech recognition processor 150 recognizes the input speech as a "call stop" command (433). If the control unit 100 recognizes the command as a "call stop" command, the control unit 100 terminates the call during the call in step 434, performs a normal call release procedure with the radio base station, and returns to the reception standby state. move on.

On the other hand, at step 450 in FIG.
0 determines whether the voice recognition processor 150 recognizes a specific character ("abbreviated key") command for the speed dial function, and if it recognizes the "abbreviated key" command, abbreviated dial status. (451)
Thereafter, it is determined whether or not the character is recognized (45).
2). At this time, if a character and a send command (or a character, a number, and a send command) that are continuously input within a predetermined time together with the specific character are recognized by voice (steps 452 to 452).
461), the registration number stored in the third memory of the speech recognition processor is searched, and the registration number is reproduced (462, 463).

In the reproduced state, step 42 in FIG.
When the command returns to step 7 and the send command is recognized again, the telephone number corresponding to the recognized character is automatically shortened and dialed.

The operation of such shortened dialing can be in various ways.

For example, the first is a number to be inputted by voice (eg, Ichi, Ni, San, Shi, Go, Roku, Nana, Send)
Is recognized as a telephone number, and the telephone number (123-4567) corresponding to the numeral number is dialed.

Second, when a subscriber name or trade name is registered by voice and registered with the subscriber name or trade name associated with a telephone number, words (such as the registered name or trade name) are registered. For example, if Hong Kong East Send is recognized, the corresponding telephone number (in the case that "Koichi East" is registered at 123-4567, becomes 123-4567) is retrieved and reproduced, and the "Send" command is continued. If it recognizes, it will automatically dial.

Thirdly, when the area name is registered by voice and the area number of the area name is registered, words such as the registered area name (eg, Busan) are recognized by voice and the corresponding area name is registered. If the area code (“Busan” is registered as 051, it becomes 051) is searched and played, and if the “send” command is subsequently recognized, the area number is manually entered into the telephone number ( Or automatically dialed with the registered phone number).

Fourth, voice registration is performed for major cities in the world, and the area code, country code, and international telephone connection number for each communication carrier are registered together. If a word (eg, SAN JOSE) is recognized, the area code and country code of the registered city and the international telephone connection number (eg, 001 (telecommunication carrier), 01 (US state No.), 408 (SAN JOSE area code)) and play them back. If the "send" command continues to be recognized, the international telephone connection number, country code, and The area code of the city is automatically dialed together with the manually entered telephone number (or registered telephone number).

Fifth, the related department signatures registered by the key phone and registered by the extension number are registered by voice, and the extension numbers of the concerned departments are registered together. If the word corresponding to (registered as 123) is recognized by voice, the extension number is searched for and reproduced, and if the "send" command is subsequently recognized, the area code is automatically dialed.

The operation of the mobile phone through the simple voice command as described above allows the use of the mobile phone while driving the car and the usual pressing of the button of the mobile phone to promptly dial. Alternatively, the convenience of the user can be improved when a blind person calls.

[0083]

As described above, the present invention relates to a conventional portable telephone system (handset and hands-free socket).
In addition, a voice processor and an interface circuit embodying a simple logical operation are additionally provided, so that the cellular phone can be operated by voice without changing the conventional circuit.

Also, according to the present invention, the power consumption can be reduced by selectively activating or deactivating the voice recognition processor when the power consumption to be used is the internal battery of the handset in accordance with the trend of miniaturization of the mobile phone. It can be plugged into a hands-free socket, and if it is an external power source, the voice recognition processor can be activated unconditionally to prevent the distraction or accident caused by the key operation of the mobile phone when driving the car.

The present invention can be applied to general home telephones and public telephones.

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments of the present invention at all, and may be made without departing from the scope of the invention.
Of course, it can be implemented in various modes.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a mobile phone system of the present invention.

FIG. 2 is a state diagram illustrating a voice recognition operation method for a mobile phone according to the present invention.

FIG. 3 is a flowchart illustrating a voice recognition operation method for a mobile phone according to the present invention.

FIG. 4 is a flowchart for explaining another embodiment of the voice recognition operation method according to the present invention.

FIG. 5 is a flowchart for explaining another embodiment of the voice recognition operation method according to the present invention.

[Explanation of symbols]

10 control unit, 150 voice recognition processor, 160
Handset key input section, 170 cover switch.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04B 7/26 109H F-term (Reference) 5K027 AA11 BB02 EE11 FF01 FF22 HH20 HH21 5K036 AA07 BB01 DD01 DD17 DD25 DD46 JJ03 JJ10 JJ13 KK07 KK09 5K067 AA34 AA43 BB04 DD16 DD27 DD54 FF07 FF27 FF38 GG01 GG11 GG21 HH22 HH23 KK13

Claims (22)

[Claims] (A) In a reception standby state, when a power-on command is voice-recognized, a step is switched to a call waiting state, and a speech command is voice-recognized while a bell ringing sound is detected. And (b) in a call waiting state, when a telephone number and a send command continuously input within a predetermined time are recognized by voice, the telephone number is dialed and the call is made. And (c) ending the call and switching to a reception standby state when a call stop command is recognized in the call state. Method. 2. The method according to claim 1, further comprising the step of: when the stop clear command is recognized by voice in the step (b), switching back to a call waiting state. . And (b) recognizing and storing the entire telephone number, reproducing and outputting the stored telephone number, and recognizing a stop clear command after reproducing and outputting. If there is an error in the input phone number, delete the entire stored phone number and return to the call waiting state.If the send command is recognized after playback and output, it is stored. The method of claim 1, further comprising dialing the phone number. 4. The voice recognition processor according to an output signal of a voice dialing on / off switch when the power consumption of the voice recognition processor performing the steps (a) to (c) is used by an internal power supply. The method according to claim 1, wherein on / off of the mobile phone is controlled. 5. When the power consumption of the speech recognition processor performing the steps (a) to (c) is used as an external power supply of the hands-free socket, the speech recognition processor is continuously driven by the external power supply. The method according to claim 1, wherein the voice recognition operation is performed. 6. The reference voice pattern stored in the voice recognition processor performing the steps (a) to (c) includes inputting a voice command and a number and a symbol of a telephone number in a user's personal voice. The method according to claim 1, wherein the method is performed by receiving and registering. 7. The reference voice pattern stored in the voice recognition processor performing the steps (a) to (c) is a voice command, a number and a symbol obtained by an average pronunciation of many people. The method of claim 1, wherein the voice recognition operation is performed on a mobile phone. 8. In the step (b), in a call waiting state, a character and a send command which are continuously input together with a specific character designated in advance for a speed dial function within a predetermined time are recognized. The method of claim 1, further comprising: searching for a telephone number corresponding to the recognized character, automatically dialing the searched telephone number, and switching to a busy state. Voice recognition operation method for mobile phones. 9. The voice of the mobile phone according to claim 8, wherein the characters recognizable in the call waiting state include a subscriber name, a trade name, a domestic and foreign region name, and a related department signature. Recognition operation method. 10. The abbreviated dialing step in the outgoing call waiting state includes the steps of: connecting an international telephone connection number classified by a communication carrier; a country code classified by each country; and a domestic and foreign region number and a subscriber telephone number. The method of claim 9, wherein dialing is performed. 11. A method for measuring a similarity between an input pattern obtained by extracting a feature parameter from an input voice command and a voice signal including a telephone number and a preset reference pattern, and determining an input voice. A speech recognition processor for recognizing, the speech recognition processor, a handset key input unit,
An interface unit that is connected to the cover switch and that recognizes an output signal value of the cover switch or the handset key input unit by a speech recognition processor and logically associates the corresponding signal value with a corresponding signal value; And a control unit for controlling overall operation of the mobile phone related to storing and deleting the recognized phone number, dialing the phone number, starting and ending the call. Recognition system. 12. The voice recognition system of claim 11, wherein the voice recognition processor is driven by selectively receiving an internal battery power by turning on / off a selection switch. . 13. The speech recognition processor according to claim 11, wherein the speech recognition processor is driven by being supplied with power from an external power source coming in through a hands-free socket.
A speech recognition system for a mobile phone according to claim 1. 14. The speech recognition processor according to claim 11, wherein the speech recognition processor includes a memory storing a reference pattern generated by inputting and receiving the voice command and the telephone number in a user's personal voice. A speech recognition system for a mobile phone as described. 15. The voice recognition processor according to claim 11, further comprising a memory storing a voice command obtained from an average pronunciation of many people and a reference pattern generated by a telephone number. A speech recognition system for a mobile phone according to claim 1. 16. The speech recognition processor according to claim 1, further comprising: extracting a feature parameter from a voice signal including a voice command, a pre-registered character and a telephone number, and acquiring a feature parameter from the voice signal. The speech recognition system for a mobile phone according to claim 11, wherein the speech is recognized by measuring the degree. 17. The voice recognition processor according to claim 16, further comprising means for detecting a connection state of an external power supply / an internal battery power supply, and operating only when the external power supply is connected. A speech recognition system for a mobile phone as described. 18. The speech recognition processor, wherein the speech recognition processor is installed outside the telephone through an interface unit extended by a hands-free socket, connected to the control unit, and driven by receiving power from an external power supply. The voice recognition system for a mobile phone according to claim 16, wherein 19. A hands-free socket having a voice recognition processor for recognizing a voice command and a telephone number inputted through a microphone, and being connected to the hands-free socket by wire / wireless to communicate with each other, A handset that receives the voice command recognized by the voice recognition processor and the telephone number, and performs the operations related to telephone number dialing, call initiation and call termination according to the command; A voice recognition system for a mobile phone. 20. The speech recognition processor according to claim 19, further comprising a memory storing a reference pattern generated by receiving the input of the voice command and the telephone number in a user's personal voice. A speech recognition system for a mobile phone according to claim 1. 21. The speech recognition processor according to claim 21, further comprising a memory storing a reference pattern generated from the voice command and a telephone number obtained from an average pronunciation of many people. 20. A speech recognition system for a mobile phone according to claim 19. 22. The speech recognition processor pre-registers a subscriber name and a mutual name, a domestic and foreign region name, and a related department name, and connects the international call classified by a communication carrier to the pre-registered character. 20. The voice recognition system for a mobile phone according to claim 19, further comprising a memory unit for storing the country code, the country code classified by each country, the domestic and foreign area code, and the subscriber's telephone number in correspondence with each other. .