KR20010000595A - Mobile phone controlled by interactive speech and control method thereof - Google Patents

Abstract

PURPOSE: A method for controlling a mobile telephone terminal capable of interactive voice control, is provided to select and control each menu of the mobile telephone terminal through interaction between a user and the terminal without searching the menu through key input each time. CONSTITUTION: Whether to select a registration mode or a dialing mode is asked to a user. If the user selects the registration mode for registering the name of a receiver and a telephone number, a voice announcement is performed to request a voice input of the name of a telephone number registration object. If the name of the object is inputted in a voice, voice data compressed and encoded in a codec and a vocoder are stored in a memory for play in voice dialing. A terminal voice-announces a request for inputting the name of the telephone number registration object with text. The user inputs the name of the telephone number registration object with text. A controller generates a voice recognition model by using the inputted text data, and voice-announces the input of the telephone number for the object in a voice. If the user inputs the telephone number of the registration object in a voice, the inputted telephone number is voice-recognized with numerical tones and stored in the memory for voice dialing. Whether to end the registration is asked. If the user selects ending, the registration is ended. If the user selects continuous registration, the step requesting a voice input of the name of a telephone number registration object is returned.

Description

Mobile phone terminal with interactive voice control and control method {MOBILE PHONE CONTROLLED BY INTERACTIVE SPEECH AND CONTROL METHOD THEREOF}

The present invention relates to a mobile phone terminal capable of interactive voice control using a speaker-independent variable vocabulary speech recognizer and a voice synthesizer and a control method thereof, and more specifically, to each function such as menu selection by voice input and output between a user and a terminal. The present invention relates to a mobile telephone terminal capable of selecting and controlling the same and a control method thereof.

FIG. 1 exemplarily illustrates a control item of a mobile phone terminal in which a user controls the menu method in a general mobile phone terminal.

Conventionally, mobile phone terminals have been provided with various functions for users to select and use according to their needs. In order to select and control these functions, a keypad input method is used. For example, a user can search for a phone number by clicking on a keypad associated with personal information management, then clicking on a phone book registration keypad and finally clicking on a keypad related to phone book search. However, the keypad input method adopts the Top-Down method, which goes from the upper menu to the lower menu, so that it is very inconvenient for the user because there are times when several keypad inputs are required to select one function. The state is real.

In particular, the keypad is not easy to use when the keypad is input while driving or walking, and it may cause a safety accident when you take your eyes off the front to control the terminal. to be.

In addition, in order to select and control a specific function, a conventional mobile phone terminal displays a menu on a liquid crystal display (LCD) window and allows a user to select the menu. Getting drunk. Therefore, it is necessary to know in which menu the function to be finally selected, so the user feels very troublesome and inconvenient unless the user is familiar with the user's manual. In addition, since these menus are displayed as text only in the LCD window, it is practically impossible to control in a situation where a low vision person or a terminal cannot be seen at a close range.

Conventionally, some mobile phone terminals have been applied to the voice recognition method, but the voice recognition method applied here is a dynamic time warping (DTW) method, which requires a learning process in the terminal. In other words, there is a hassle that a user needs to directly record a target word or vocabulary to be recognized, and the recognition parameter is extracted using the recorded voice data. For this reason, the voice recognizer was used only for the voice dialing function.

Due to the above problems, various functions currently provided by the mobile phone terminal are acting as a factor of using only a small part from the user's perspective.

The present invention has been made to solve the above-described problems of the user control has a conventional mobile phone terminal, the purpose of the user and the user to communicate with each other between the terminal and the user, without having to go through the menu of each step through the key operation The present invention provides a controllable mobile phone terminal and a control method thereof.

1 exemplarily illustrates a control item of a mobile phone terminal, which is controlled to be controlled by a user in a menu method in a conventional mobile phone terminal.

2 is a block diagram schematically illustrating the components of an interactive voice dialing control system of a mobile phone terminal capable of interactive voice control according to the present invention.

3 and 4 are flowcharts illustrating an algorithm of an interactive voice dialing method in an interactive voice control method of a mobile phone terminal according to the present invention.

* Description of the symbols for the main parts of the drawings *

10: microphone 20: codec 30: vocoder

40: mobile phone terminal keypad 50: control unit 60: user voice storage memory

70: guide voice storage memory 80: liquid crystal display 90: speaker

The mobile phone terminal capable of interactive voice control according to the present invention recognizes a user's control command and outputs a code necessary for controlling the mobile phone terminal, and a user voice input means for allowing a user to input voice to the voice recognizer. And a voice synthesizer for sounding the user's various instructions required to perform a user control command input through the voice recognizer, a terminal voice output means for outputting the synthesized voice from the voice synthesizer, a name of the receiver, and the like. And a guide voice storage memory for storing voice data to be output by the voice synthesizer.

The speech recognizer may be a variable vocabulary speech recognizer capable of performing the same command by recognizing a certain category of vocabulary with the same meaning.

It is preferable that the speech synthesizer uses a recording editing type speech synthesis. In this way, the voice guided by the voice synthesizer can be changed into various voices (voices of other speakers) according to the user's request.

The interactive voice control method of a mobile phone terminal according to the present invention comprises the steps of: (a) outputting a voice request for a voice command to a user; (b) a voice command for controlling the mobile phone through a user voice input means; Receiving the input, (c) recognizing the input voice and outputting a code for controlling the mobile telephone terminal; (d) if the additional voice command requires additional detailed voice commands, the detailed voice commands It is configured to include the step of outputting a voice requesting the guidance.

The interactive voice control method may be implemented as follows.

The sentence below [terminal #] is a synthesized voice guidance message guided by the terminal, the sentence below [user #] is the content spoken by the user.

[Terminal # 1] What is your desired service?

[User # 1] Reserved Call

[Terminal # 2] Please state the telephone number you wish to call.

[User # 2] Young, Lee, Seven, Seven, Seven, Three, Three Spirits (02-3775-3330)

[Terminal # 3] What time do you want to call?

[User # 3] Nine Thirty (09:30)

[Terminal # 4] At nine thirty, thirty three thirty seven seventy five five thirty three thirty three hundred thirty calls.

Among the interactive voice control methods of the mobile phone terminal according to the present invention, in particular, a dialing control method (e) receives a recipient name and a corresponding phone number by voice and maps the corresponding phone number to a recipient name to a user voice storage memory. Step of storing in advance, (f) receiving a voice input of the name of the recipient and the phone number input in the step (e) through a key input to create a voice recognition model, mapping the characters to the corresponding recipient name to the user voice Storing in the storage memory in advance, (g) receiving a name of the receiver through a user voice input means, (h) recognizing the input name, and searching the user voice storage memory for corresponding name data (I) using the terminal voice output means as a voice inputted by the user at the beginning of the searched name and telephone number; And (j) confirming and dialing the recipient of the user.

If the name and phone number searched and outputted in step (i) are two or more, in step (j), the user's receiver selection is input by voice.

By adopting the voice dialing control method, the possibility of dialing a wrong number can be blocked, and a quick voice dialing function can be implemented.

That is, by including the receiver's name in the initial recognition target vocabulary of the voice recognizer, the receiver can be dialed as soon as the receiver's name is input in the hook-on state.

Hereinafter, with reference to the accompanying drawings an embodiment of a mobile phone terminal and a control method capable of interactive voice control according to the present invention will be described in detail.

For convenience, among the user control of various mobile telephone terminals, in particular, dialing control will be exemplarily described.

2 is a block diagram schematically illustrating the components of an interactive voice dialing control system of a mobile phone terminal capable of interactive voice control according to the present invention.

When a user registers a phone number for voice dialing, first, the name or business name of the receiver is uttered through the microphone 10, and the input voice data is changed into a PCM (Pulse Code Modulation) signal by the codec 20. Next, it is compressed into a voice coder (QCELP) signal through the vocoder 30 and then output to the controller 40. The controller performs a role of a voice recognizer and a voice synthesizer while executing a voice recognition program and a voice synthesis program. The input voice data is stored in the memory 60 to output the recognition result as voice in the voice dialing mode. And when a character such as speech content is input using the keypad 40, the controller 40 creates a model for speech recognition thereof, and the text content is stored together with the speech content in the memory 60. When the user registers the phone number to be registered, the result is stored in the memory through digital voice recognition. The control unit 10 is implemented as a microprocessor that controls the overall operation and performs a voice synthesis program of the voice recognition and recording editing.

The following exemplarily illustrates a conversation between a receiver and a terminal that can be used in the above-described receiver phone number registration mode.

[Terminal # 1] What is your desired service?

[User # 1] Phone Number Registration.

[Terminal # 2] phone number registration service. Please state the recipient's name or business name.

[User # 2] Act Valley

[Terminal # 3] Please enter the valid text of the VALLEY as mentioned.

[User # 3] (Enter ActValley as text.)

[Terminal # 4] Please tell me your phone number.

[User # 4] Young, Lee, Seven, Seven, Seven, Three, Three Spirits (02-3775-3330)

In the above example, [terminal #] is a voice that is guided through the speaker 70 by synthesizing the voice of the voice stored in the memory 70 for voice guidance according to the operation of the controller 40. This is made possible by recording the corresponding voice data so that the voice guided by the user's request can be changed into the voice of another speaker.

In the voice dialing mode, when a user speaks a person's name or a business name, the controller 50 detects a voice section and extracts feature parameters from the voice signal passing through the codec 20 and the vocoder 30, and extracts a feature parameter. Speech recognition is performed by comparison with the recognition model recorded in (60). The recognition result reproduces the voice spoken by the user at the time of registering the phone number, that is, the receiver name or the business name through the speaker 90, and simultaneously displays the text corresponding to the inputted text on the liquid crystal display 80 window.

If the recognition result is two or more, the contents are displayed on both the speaker 90 and the liquid crystal display 80, and the voice dialing is attempted by re-recognizing again through the voice recognition.

By doing so, the user can check whether the receiver to be called is correctly recognized by voice output during voice dialing.

The following is an example of a conversation between a user and a terminal that can be used in the voice dialing mode.

[Terminal # 1] What is your desired service?

You have selected [user # 1] Make a call [terminal # 2]. Please state the name or business name of the person you wish to call.

[User # 2] Act Valley

[Terminal # 3] There are 2 results. The first act valley is Samsamsamyoung in Samchilchio Guk. The second actbelly is seventy-three on Yeongi. Please tell me where you want to call.

[USER # 3] NO.1 (OR FIRST)

[Terminal # 4] I am calling Samshil Chil-guk at Samsamsamyoung.

As shown in the above example, in the case of two or more voice recognition results, as shown in [user # 3], the voice dialing is performed again to finally dial the number to be called by the user.

In the current mobile phone terminal, when a phone number is registered, a specific number is stored in the recipient's related contents, and when dialing, the stored number is clicked to make a call immediately.

In the present invention, it is possible to implement a quick voice dialing function that enables such a function by voice. In other words, in the process shown as an example of voice dialing, when the user utters a receiver who wants to make a telephone call, such as "[User # 2] Act Valley", the name or business name of the receiver registered in the telephone number is immediately connected. This is possible by including the voice recognition in the initial recognition target vocabulary.

In addition, the user may perform one operation after recognizing the same vocabulary with respect to various responses having the same meaning spoken by the user in the dialogue process between the user and the terminal. In other words, even if the user responds with "Yes", "Yes", "Yes", the mobile phone terminal is implemented to recognize each and take the same action for all of them. This is possible by adopting the variable vocabulary voice recognizer as the food recognizer of the mobile phone terminal capable of interactive voice control according to the present invention.

3 and 4 are flowcharts illustrating an algorithm of an interactive voice dialing method in an interactive voice control method of a mobile phone terminal according to the present invention.

3 shows the algorithm of steps (e) and (f) in more detail.

The user is asked whether to select a registration mode or a dialing mode (200), and when the user selects a registration mode for registering a recipient name and a phone number, a voice guidance requesting a voice number registration target name voice input is performed (202). . When voice input of the name of the telephone number registration target (203), the voice data compressed by the codec and vocoder are stored in the memory for playback during voice dialing.

Again, the terminal provides a voice guidance requesting the name of the phone number registration target book as text (205), and receives the name text of the phone number registration target user from the user (206). The control unit 50 generates a voice recognition model using the input text data (208), and voice guides the user to voice input the phone number of the subject (208).

When the user inputs the phone number of the person to be registered (209), it stores it in a memory for voice dialing after voice recognition of the numeric voice (210). If the user is asked whether to end the registration (211) and selects the end of the registration, the registration is terminated. If the registration is to be continued, the step 202 is performed again.

Figure 4 shows an embodiment of an algorithm implementing the steps (g) to (j).

In step 200, when the user selects the dialing mode, the voice receiver 301 inputs a name of a receiver to be called according to the voice guidance 301.

The voice recognizer extracts a voice recognition parameter (303), recognizes a voice (304), retrieves and outputs a name matching the recognition result from the user voice storage memory (305), and provides a voice guidance requesting a user's response ( 306).

As described above, if there are two or more recognition results, the number is given according to the output order, that is, 1 for the first output name, or 2 or 2 for the first output name. Dog selection (309), a telephone call is attempted (308).

According to the present invention, the following effects can be obtained.

First, by selecting and controlling various menus through a dialogue between the user and the terminal, it is possible to secure convenience and stability when moving.

Second, in the phone number registration mode, the voice of the receiver's name is stored, and the voice recognition results in recognizing the receiver's name stored in memory through the speaker. In addition, it is possible to prevent a call to another number due to misrecognition. And considering that the safety accidents caused by the use of a mobile phone during driving or walking due to the most attempts to make a phone call, the present invention has the effect of preventing such a safety accident in advance.

Third, the voice guided voice in the interactive mobile phone terminal of the present invention can be changed to a variety of voices if necessary by the user, thereby eliminating the nausea of the guide sound due to continuous use.

Fourth, in the interactive method of multi-step voice dialing, the voice caller's name or company name can be connected to a call immediately when the user's name or business name is called. Can give

Fifth, the terminal recognizes the voice answers of a plurality of users commonly used in the same meaning as a vocabulary having the same meaning, thereby adding convenience.

Claims (5)

A voice recognizer for recognizing a control command of a user and outputting a code necessary for controlling the mobile telephone terminal; User voice input means for allowing a user to input voice to the voice recognizer; A voice synthesizer for providing a voice to the user with various guides necessary to perform a user control command input through the voice recognizer; Terminal speech output means for outputting speech synthesized by the speech synthesizer; A user voice storage memory for storing voice data of a user, such as a name of a recipient and a list of telephone numbers mapped; A guide voice storage memory for storing voice data to be output by the voice synthesizer; Mobile phone terminal capable of interactive voice control, characterized in that configured to include. The mobile phone terminal of claim 1, wherein the voice recognizer is a variable vocabulary voice recognizer capable of performing a same command by recognizing a certain category of vocabulary with the same meaning. The mobile phone terminal according to claim 1, wherein the voice synthesizer is a voice synthesizer of a recording editing method which can change the guided voice into various voices according to a user's request. (a) outputting a voice prompt for requesting a user to make a voice command; (b) receiving a voice command of a user for controlling the mobile phone through a user voice input means; (c) recognizing the input voice and outputting a code necessary for controlling the mobile telephone terminal; (d) if an additional detailed voice command is required for the input voice command, outputting a voice request for the detailed voice command; Interactive voice control method of a mobile phone terminal, characterized in that comprising a. (e) receiving a receiver name and a corresponding telephone number by voice and mapping the telephone number to the receiver name and storing the telephone number in advance in a user voice storage memory; (f) Create a voice recognition model by receiving text input from the name of the receiver and the phone number input in step (e) through key input, and mapping the text to the corresponding receiver name to store it in the user's voice storage memory in advance. Doing; (g) receiving a name of a receiver through a user voice input means; (h) recognizing the input name and retrieving corresponding name data by searching the user voice storage memory; (i) outputting the retrieved name and telephone number through the terminal voice output means as voice input by the user initially; And (j) confirming and dialing the recipient of the user; Interactive dialing control method of a mobile phone terminal, characterized in that comprising a.