JPH10200617A - Portable telephone set with voice-recognition function - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a small-sized telephone set with superior carrying performance and few operation buttons. SOLUTION: When a 'voice' button is depressed in an incoming call waiting state, the processing transits to a voice repertory dial. A voice-recognition section 110 recognizes a received voice and reports a registration number object of a destination to a control section 107 by referring to a dictionary memory 111. When the 'voice' button is depressed twice in the incoming call waiting state, the processing transits to a voice dial state. The telephone number is entered digit by digit through the entered voice or the depression of a 'function' button 105. When a voice input is received, the voice-recognition section 110 recognizes the voice signal and reports a numeral code object to the control section 107 by referring to the dictionary memory 111. The control section 107 displays the numeral and moves a cursors to a next digit. In the case of depression of the 'function' button 105, a numeral is sequentially displayed during the depression, and the display is confirmed when the depression is stopped to move the cursor to the next digit. When a 'speech' button is depressed, dialing is made by the registration number or by a numeral string through one-digit entry, so as to attain a speech.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention belongs to the technical field of a telephone with a voice recognition function that can be carried and used.

[0002]

2. Description of the Related Art Conventionally, there is a so-called voice dialing technique for inputting a dial number by voice recognition as a technique for eliminating the need to press a dial button on a telephone. However, if the surrounding noise is loud or if a dictionary necessary for recognition of the input voice is not properly created, voice recognition becomes difficult, and an alternative input means is required. In the related art, when voice recognition is difficult, an input is made using a dial button on a telephone.

[0003]

In the above prior art,
Even if a voice dial function is provided, it is necessary to mount a separate dial button as a countermeasure when voice recognition is difficult, and this dial button has been a factor limiting the miniaturization of the telephone.

[0004] It is an object of the present invention to provide a small and highly portable telephone having a small number of operation buttons.

[0005]

SUMMARY OF THE INVENTION To achieve the above object, the present invention relates to a telephone for making a dial call using voice recognition means, comprising: a transmitting means for receiving a call, a receiving means, Means for starting and ending the call, means for inputting the dial number of the call destination, and means for confirming the input dial number. The input of the digits of the dial number is performed by the voice recognition means. Or 1
It is characterized in that the operation is performed by pressing one of the operation button switches.

In addition to the above-mentioned portable telephone with a voice recognition function, a means for inputting and registering a destination voice of a dial destination, a means for inputting and registering a dial number corresponding to the destination voice, Means for confirming the entered dial number, means for recognizing the registered address voice from the input voice, and means for transmitting a dial number corresponding to the recognized address, and registering the dial number. Is input by voice recognition by the voice recognition means or by pressing one of the operation button switches.

Further, the voice recognition means or the means for confirming the dial number inputted by pressing the one operation button switch is a display or an audible sound reproducer.

In the present invention, the dial button is not required by the voice recognition means, and the dial input can be performed by operating at least one button even when voice recognition is difficult, so that the number of operation buttons is small. To provide a small, highly portable telephone.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<< Outline of Configuration >> FIG. 1 is a block diagram of an embodiment of a portable telephone having a voice recognition function according to the present invention, wherein 101 is a transmitter, 102 is a receiver, 103
Is a button for mainly performing a call start and call end operation (hereinafter referred to as a "call" button), 104 is a button for mainly performing a voice input operation (hereinafter referred to as a "voice" button),
Reference numeral 105 denotes a button for performing various functions corresponding to each state (hereinafter referred to as “function” button), 106 denotes a display, 107 denotes a control unit, 108 denotes a wireless transmission / reception unit, 109 denotes a communication circuit unit, and 110 Is a voice recognition unit, and 111 is a dictionary memory. Transmitter 101, receiver 102, wireless transceiver 1
08, the communication circuit unit 109 is a part necessary for the configuration of a general portable cordless telephone. Here the transmitter 10
1 and the receiver 102 can be implemented in either a loudspeaker type or an earset type. "Call" button 103, "Audio" button 104, "Function" button 1
05, display 106, control unit 107, voice recognition unit 11
0, the dictionary memory 111 will be described later.

<< Explanation of Operation >> Next, an operation example of this embodiment will be described. FIG. 2 is a flowchart showing the overall operation flow of the embodiment, and reference numerals 201 to 219 denote processing and operation steps. Note that the circles A in the figure indicate that the processing or operation continues between them.

(1) Power on / off When the power is off, the "call" button is pressed and held down for 5 seconds, for example, to turn on the power (step 201). For example, when the button is kept pressed for 5 seconds, the power is turned off (step 219).

(2) Starting and ending the call After the power is turned on, an incoming call waiting state is set (step 20).
2). If there is an incoming call (step 203),
Answering the incoming call by pressing the "call" button (step 20)
4), the call starts (step 215). In addition, in the incoming call waiting state, pressing the "voice" button shifts to voice repertoire dialing (step 206), and pressing the "voice" button again shifts to voice dialing (step 209). Voice Repertoire Dial Status (Step 21)
0) and in the voice dialing state (step 211), the dialing is performed by pressing the "call" button after the dial number is determined (step 214). After the call is started (step 215), the call is terminated by pressing the "call" button again (step 216), and the apparatus enters the incoming call waiting state (step 21).
8).

(3) Voice Repertoire Dialing As described above, when the "voice" button is pressed in the incoming call waiting state, a transition is made to the voice repertoire redial state (step 210). An operation example of the voice repertoire dial will be described with reference to FIG. Upon entering the voice repertoire dialing state, an address utterance instruction is displayed (step 30).
1). Following the instruction, when the address registered in advance is uttered (step 302), the recognition result is displayed after the utterance is completed (step 303). If the displayed result is the address to be transmitted this time, the voice repertoire dialing is completed (step 304), and the dialing can be performed by pressing the "call" button (step 214). On the other hand, if the address information to be transmitted is not displayed in step 303, one of the following two operations can be performed. The first is an operation of pressing a “function” button (step 305). At this time, the next candidate of the recognition result is sequentially displayed each time the button is pressed (step 307). The second is utterance again (step 306). Here, after the utterance, the recognition result is displayed (step 308). Step 307 or Step 3
If the address information to be transmitted is displayed at 08, the voice repertoire dialing is completed (step 309),
Subsequently, dialing can be performed by pressing the "call" button (step 214).

(4) Voice dialing In the incoming call waiting state, pressing the "voice" button twice in succession shifts to the voice dialing state (step 21).
1). An operation example of the voice dial will be described with reference to FIG.
Immediately after entering the voice dialing state, the state shifts to the one-digit input state (step 401), and ends when all the digits have been input (step 402). Thereafter, by pressing the "call" button, a call can be made with the input dial number (step 214).

(5) Single Digit Input A single digit input in the single digit input state will be described with reference to an operation example of FIG. 5 and a display example of FIG. First, FIG. 6A shows a display example of waiting for input in a single digit input state. In this figure, input of the first digit is awaited, and the cursor flashes at the digit (step 501). Next, in this state, one of two operations is possible. The first is the utterance of dialed digits (step 502). Here, after the utterance, the dial number of the recognition result is displayed in the digit (step 5).
04), the cursor moves to the next digit and blinks (step 506). An example of the display is shown in FIG. The second is an operation of pressing a “function” button (steps 503 and 5).
05). At this time, when the "function" button is continuously pressed, a number is displayed at the cursor designated digit, and the display is sequentially switched while the button is pressed (step 503). FIG. 6B shows an example of the display. For example, assuming that the display is switched to “0”, “1”, “2”,... Every 0.5 seconds by continuous pressing, “5” in FIG. It will be displayed between 2.5 seconds and 3.0 seconds later. If the user stops pressing the "function" button during this time, the number of the digit is determined (step 505).
The cursor waiting for input moves to the next digit and blinks (step 506). An example of the display is shown in FIG. If the user cancels the input contents and wants to input the digit again (step 507), the user presses the "voice" button and the "function" button simultaneously to delete the number displayed in the digit (step 508). By this operation, the cursor moves back by one digit (step 509), and blinks at that position (step 501). By the above method, it is possible to input one digit at a time, and this is repeated for all digits (steps 402 to 401).
Loop), dial number input is completed. Thereafter, a dial call can be made by pressing the "call" button (step 214).

(6) Dial Digit Registration An example of an operation of registering an operator's voice for dialed digits and creating a dictionary prior to the operation of the voice repertoire dial or the voice dial will be described with reference to FIG. In FIG. 2, the state of FIG. 7 is started by pressing two buttons of “voice” and “function” simultaneously (step 207). In the dial digit registration state, the dial digit “n” is pressed by pressing the “voice” button.
An utterance instruction is displayed (here, n = 0, 1, 2,...).
(Step 701). Following this instruction, "n"
(Step 702). Subsequently, when the "function" button is pressed, the input is confirmed by the utterance (step 704). On the other hand, when re-uttering, the same numeral "n" is re-uttered (steps 703 to 70).
2). When the above operation is repeated for all dialed digits (step 705), if the "call" button is pressed, the process ends (step 217).

(7) Voice Repertoire Redial Registration Voice repertoire redial registration is to register an address voice and a dial number in association with each other. The operation will be described with reference to FIG. Note that the circles A to C in the figure indicate that processing and operations continue between the same symbols.

In FIG. 2, the state of FIG. 8 is started by simultaneously pressing two buttons of "voice" and "call" (step 208). Step 801 in FIG. 8 shows that the registration status of the first registration content of the repertoire dial is displayed. The next step indicates that the process branches depending on the type of button to be pressed. In step 802, "voice"
Pressing the button starts repertoire registration. In this state, the address is uttered first (step 805). Vocalization is OK
If it is (step 806), the input is confirmed by pressing the "voice" button again (step 80).
7). On the other hand, the re-utterance is pressed by the "function" button (step 8).
08). When the "voice" button is pressed after the address input is confirmed, the state shifts to the dial number input state (step 809). On the other hand, when the "function" button is pressed, the address registration is terminated (step 810). This corresponds to, for example, a case where only the address is changed without changing the dial number. In the dial number input state, the one digit input of the dial number in step 811 is repeated, and when the input of all digits is completed (step 812), the input is confirmed by pressing the "voice" button (step 813).

As another operation due to the branch of the processing in FIG. 8, in step 803, the registered contents of the number are erased by simultaneously pressing two buttons of "voice" and "function". Also,
In step 804, only the registration number is updated by pressing the "function" button.

In FIG. 8, when all necessary registrations have been completed (step 815), the processing is terminated by pressing the "call" button (step 217).

<< Control of Operation >> The operation of the embodiment of the present invention has been described above. The control of each unit in FIG. 1 for realizing the operation will be described below.

(1) Power ON / OFF In the power OFF state, the control unit 107
3 is monitored. Here, if the "call" button 103 is continuously pressed for 5 seconds, each operation is enabled by turning on the power. In the power-on state, the control unit 107 performs “call”
It monitors whether the button 103, the “voice” button 104, or the “function” button 105 is pressed. When the “call” button 103 is continuously pressed for 5 seconds, the power is turned off.

(2) Start and end of call After the power is turned on, the control unit 107 controls the radio transmission / reception unit 108 to be in a call waiting state. If there is an incoming call, the press of the "call" button 103 is monitored, and the "call" button 10
3, the communication circuit unit 109 and the wireless transmission / reception unit 1
08 is set to the call start state. When the “call” button 103 is pressed in a call state, the call path is disconnected.

(3) Voice Repertoire Dial If the "voice" button 104 is pressed in the incoming call waiting state, the operation shifts to voice repertoire dialing. After the shift, the control unit 107 connects the speech signal path of the speech circuit unit 109 and the speech recognition unit 110 to enable recognition of the input speech. When a voice is input from the transmitter 101 to the voice recognition unit 110, the voice recognition unit 110 extracts features of the input voice and compares it with an address voice pattern stored in the dictionary memory 111 in advance. As a result, the control unit 107 is notified of the registration number corresponding to the address voice having the highest similarity as a recognition candidate. Here, the number of recognition candidates to be notified may be plural. The control unit 107 reads out the registered number of the notified first recognition candidate and the corresponding dial number from the dictionary memory 111, for example, and reads the registered number from the dictionary memory 111.
6 is displayed. The control unit 107 sets the voice recognition unit 110 in a voice standby state in preparation for re-speaking, and prepares a “function” button 105 for a display request of a second recognition candidate.
Monitor the press of. Further, it monitors the pressing of the "call" button 103 in preparation for a dialing request.

(4) Voice dialing When the "voice" button 104 is pressed twice in the incoming call waiting state, the state shifts to the voice dialing state. In this state, the control unit 107 blinks the cursor of the input waiting digit portion of the display unit 106. On the other hand, in order to enable recognition of the input voice, the voice signal paths of the speech circuit unit 109 and the voice recognition unit 110 are connected.

(5) Single digit input When a voice is input from the transmitter 101 to the voice recognition unit 110, the voice recognition unit 110 extracts the characteristics of the input voice and stores the numbers stored in the dictionary memory 111 in advance. The matching is performed with the voice pattern. As a result, a code corresponding to the numerical voice pattern having the highest similarity is notified to the control unit 107 as a recognition candidate. The control unit 107 displays the number corresponding to the code in the input waiting digit portion of the display unit 106, moves the cursor to the next digit, and blinks it. In this state, the control unit 107 sets the voice recognition unit 110 in a voice input standby state in preparation for the next voice input, and furthermore, the “voice” button 104, the “function” button 105,
It monitors the pressing of three buttons of the “call” button 103.
When the two buttons of the "voice" button 104 and the "function" button 105 are simultaneously pressed, the cursor is returned by one digit, the displayed number is erased, and the cursor blinks. Also "Function"
When the button 105 is pressed, the duration of the pressing is measured, the number to be displayed is displayed when the pressing is stopped, the cursor is moved to the next digit, and the button 105 blinks. When the “call” button 103 is pressed, the communication circuit unit 109 is disconnected from the voice recognition unit 110 and connected to the wireless transmission / reception unit 108 in order to shift to a dial transmission operation using the already input digit string. The transmission / reception unit 108 is instructed to send a dial.

The above-described waiting for voice input and monitoring of pressing of three buttons will be described in further detail.
Specifically, it is necessary to invalidate the button press during the voice input and to invalidate the voice input during the button press. First, a method for invalidating button press during voice input will be described. After the control unit 107 sets the voice recognition unit 110 to the voice input standby state, and immediately after the voice recognition unit 110 detects the start of voice input, the control unit 107
To notify voice input. Normally, the voice recognition unit 110
This notification can be easily made because the presence or absence of the voice input can be detected every 20 ms or less. If there is this voice input notification, the control unit 107 may invalidate subsequent button presses. On the other hand, if the notification of the recognition candidate is received from the voice recognition unit 110 after the voice input is completed, and the display corresponding to the notification is completed, the button press can be made valid again thereafter.
Next, a method of invalidating the voice input while the button is pressed will be described. If the button press is started after the control unit 107 sets the voice recognition unit 110 to the voice input standby state, the control unit 107 notifies the voice recognition unit 110 of the release of the voice input standby, and releases the pressed button. What is necessary is just to shift to the process corresponding to the type.

(6) Dial Digit Registration If two buttons, the "voice" button 104 and the "function" button 105, are pressed simultaneously in the incoming call waiting state, the process proceeds to dial digit voice registration. After the shift, the control unit 107 connects the speech signal path of the speech circuit unit 109 and the speech recognition unit 110 so that the input speech can be registered. Next, when the “voice” button 104 is pressed, the control unit 107
6 displays a numeric utterance instruction message. Thereafter, when a voice is input from the transmitter 101 to the voice recognition unit 110, the voice recognition unit 110 performs feature extraction of the input voice and writes a voice pattern into, for example, a memory buffer area. After this, if the “function” button 105 is pressed, the input is confirmed,
The voice pattern is transferred from the memory buffer area to the dictionary memory 11.
1 is transferred and stored in the area corresponding to the number. on the other hand,
When the voice is input again, the voice recognition unit 110 performs the feature extraction of the input voice and writes the voice pattern into the memory buffer area again.

(7) Voice Repertoire Dial Registration If two buttons, the "voice" button 104 and the "call" button 103, are simultaneously pressed in the incoming call waiting state, the process proceeds to voice repertoire dial registration. After the shift, the control unit 107 connects the speech signal path of the speech circuit unit 109 and the speech recognition unit 110 so that the input speech can be registered. Next, when the “voice” button 104 is pressed, the control unit 107 displays an address utterance instruction message on the display unit 106. Thereafter, when a voice is input from the transmitter 101 to the voice recognition unit 110, the voice recognition unit 110 performs feature extraction of the input voice and writes a voice pattern into, for example, a memory buffer area. Next, the control unit 107 monitors pressing of the “voice” button 104 and the “function” button 105. Here, if the “voice” button 104 is pressed, the input is determined, and the voice pattern is transferred from the buffer area to the dictionary memory 111 corresponding to the number and stored. On the other hand, when the “function” button 105 is pressed, the control unit 107 instructs re-input of the address. If the "voice" button 104 is pressed again after the input is confirmed by pressing the "voice" button 104, the state shifts to the registration state of the dial number corresponding to the address. Subsequent control is the same as in the case of single digit input. Thereafter, when the "voice" button 104 is pressed again, the dial number entered so far is determined and stored in the area corresponding to the registration number in the dictionary memory 111. Control unit 107 then updates the registration number and repeats the process.

<< Method of Confirming Dial Number Input >> In the above description, it has been described that confirmation of dial number input is performed by the display unit 106. In this case, the operator can confirm with visual information, but the same can be done with auditory information. For example, the display of digit recognition candidates in voice dialing (step 504 in FIG. 5) can be realized by reproducing a voice signal from the receiver 102. Specifically, the control unit 107 converts the recognition candidate data into audio signal data. Then, the audio signal data is transmitted to the communication circuit unit 109. The voice signal data is generally a digital signal.
If means for converting the audio signal data from digital to analog is provided in 9, the converted analog audio signal is transmitted to the receiver 102 and can be heard as audible sound. Similarly, when dialing a number by pressing a button, the content can be confirmed by voice. For example, in the case of single-digit input, the numeric display is sequentially switched by successively pressing the “function” button 105 (step 50 in FIG. 5).
In order to perform the operation corresponding to 3) by voice, during continuous button press, “zero”, “one”, “d”,.
The time interval for reproducing the numeric voice may be set long enough so that the operator stops listening to the number to be input and then stops pressing the "function" button 105 so that the input can be confirmed. For example, the operator may stop pressing the “function” button 105 and reproduce the number whose input is confirmed again to notify the operator.

<< Instruction to Operator >> Although the method of confirming dial number input has been described above, the next operation instruction to the operator can also be realized by displaying a message on the display unit 106. Alternatively, it can be realized by reproducing the message with an audible audio signal by the receiver 102.

<Effects> As described above, the present embodiment is
The telephone function is realized only by having three operation buttons, voice recognition means, and confirmation means of the destination dial number. For this reason, the space required for arranging buttons and the like can be significantly reduced as compared with a telephone having conventional dial buttons. In addition, the voice repertoire dial function enables dialing by a simple operation of speaking an address. Furthermore, dial numbers can be input by either voice recognition or one-button operation, which reduces the difficulty of inputting due to an increase in ambient noise and a change in vocalization, which were disadvantages of conventional voice dialing. Dial operation is possible without inviting. As a first example, when the surrounding noise level suddenly increases in the middle of voice dialing and it becomes difficult to input subsequent dial digits with voice, it is possible to switch subsequent dial input to button operation. . As a second example, when voice input is difficult due to a change in the utterance state due to a cold or the like, dial number input can be switched to button operation.
This is because in the case of a set of words as short as a number, it is not easy to identify features between words, and the recognition rate tends to decrease due to fluctuations in utterance. As described above, according to the present invention, the voice input and the button input are combined, so that the dial operation can be reliably performed even in such a state.

[0034]

As is apparent from the above description, according to the present invention, the telephone function can be realized only by having at least one operation button, voice recognition means, and means for confirming the destination dial number. Compared with a telephone having conventional dial buttons, the space required for arranging buttons and the like can be greatly reduced.

Further, according to the present invention, a voice input and a button input can be combined to perform dial digit input either by voice recognition or one button operation, which is a disadvantage of the conventional voice dial. The dial operation can be performed without causing the input difficulty due to the increase in the ambient noise and the change in the utterance state.

Further, when the address voice and the corresponding dial number are registered in advance and the address voice is recognized from the input voice, dial transmission can be performed by a simple operation of speaking the address. .

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a flowchart illustrating an operation example in the embodiment.

FIG. 3 is a flowchart illustrating an operation example of the voice repertoire dial in FIG. 2;

FIG. 4 is a flowchart illustrating an operation example of the voice dial in FIG. 2;

FIG. 5 is a flowchart illustrating an example of the operation of single digit input in FIG. 4;

FIGS. 6A, 6B, and 6C are diagrams illustrating display examples in the operation example of single digit input in FIG. 5;

FIG. 7 is a flowchart illustrating an operation example of dial digit registration in FIG. 2;

FIG. 8 is a flowchart for explaining an operation example of voice repertoire dialing registration in FIG. 2;

[Explanation of symbols]

 101: transmitter 102: receiver 103: "call" button 104 ... "voice" button 105: "function" button 106 ... display 107 ... control unit 108 ... wireless transmission / reception unit 109 ... communication circuit unit 110 ... voice recognition unit 111 … Dictionary memory

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Yoshio Nakadai 3-19-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo Nippon Telegraph and Telephone Corporation

Claims (4)

[Claims] 1. A telephone set for making a dial call using voice recognition means, comprising: a transmitting means for receiving a call, a receiving means, a means for starting and ending the call, and a dial number of a call destination. And means for confirming the input dial number, wherein the input of the number of the dial number is performed by voice recognition by the voice recognition means or by pressing one operation button switch. A mobile phone with a voice recognition function. 2. A means for inputting and registering an address voice of a dial destination, and a means for inputting and registering a dial number corresponding to the address voice, in addition to the portable telephone having a voice recognition function according to claim 1. Means for confirming the entered dial number, means for recognizing the registered address voice from input voice, and means for transmitting a dial number corresponding to the recognized address, the dial The input of a number for registering a number is performed by voice recognition by the voice recognition means or by pressing one of the operation button switches. A portable telephone with a voice recognition function, characterized in that: 3. The display device according to claim 1, wherein the voice recognition unit or the unit for confirming the dial number input by pressing the one operation button switch is a display unit. A mobile phone with a voice recognition function as described. 4. The audible sound reproducer according to claim 1, wherein the voice recognition means or the means for confirming the dial number input by pressing the one operation button switch is an audible sound reproducer. 3. The mobile phone with a voice recognition function according to 2.