KR20040048435A - Voice-controlled television set and operating method thereof - Google Patents

Abstract

PURPOSE: A voice controlled television receiver and a voice control method are provided to recognize a viewer's voice command through a microphone to control a television receiver. CONSTITUTION: A voice controlled television receiver includes speakers(30,31), an audio amplifier(40), an internal/external microphone(50), a speaker interference sound removing device(60), and a voice recognition device(70). The speaker interference sound removing device extracts an electric signal driving the speakers to estimate an interference sound digital signal and removes the estimated signal from a signal input to the microphone to eliminate an interference sound. A voice command signal(51) passes through the speaker interference sound removing device to become a pure voice command signal(52), which is transmitted to the voice recognition device.

Description

VOICE-CONTROLLED TELEVISION SET AND OPERATING METHOD THEREOF}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a television receiver and a control method for recognizing a voice command of a viewer, and more particularly, to a speech recognition television receiver and a voice control method for solving self-sound interference generated from a television receiver speaker.

The voice recognition television receiver receives voice commands from a viewer through a microphone built in the receiver main body or an external microphone, and basic operations such as power on / off, channel change, and volume control are controlled by voice. Control techniques for voice recognition television receivers are disclosed in US Pat. No. 6,119,088, Japanese Patent No. 5,289,690, and the like.

However, in the case of the voice recognition television receiver according to the prior art, since the sound generated from the television receiver's own speaker is reflected directly or around the object and input to the microphone together with the viewer's voice, the voice command signal of the viewer is severely interrupted. . As a result, the viewer's voice command recognition rate is significantly lowered.

Accordingly, a first object of the present invention is to provide a television receiver and a control method which are controlled by recognizing a voice command of a viewer through a microphone built in or external to the main body.

The second object of the present invention is, in addition to the first object, a television receiver and a control method comprising an interference canceling device for removing a speaker generated sound of a television receiver from inside a television receiver. To provide.

1 shows an embodiment of a voice controlled television receiver equipped with a built-in or external microphone according to the present invention.

Figure 2 illustrates one embodiment of a building block for speaker interference noise cancellation of a voice controlled television receiver according to the present invention.

Figure 3 is a view showing an embodiment of the configuration of a speaker interference cancellation device according to the present invention.

4 illustrates an embodiment of a variable coefficient digital delay line filter in accordance with the present invention.

5 shows an embodiment of a filter coefficient generator in accordance with the present invention.

<Explanation of symbols for the main parts of the drawings>

10: external microphone

20: built-in microphone

30: left speaker

31: Us picker

32: TV receiver

40: amplifier

41: Speaker drive signal

42: analog-to-digital converter (A / D converter)

51: voice command signal

60: speaker interference cancellation device

62: variable coefficient digital delay line filter

63: Adder / Subtracter (Adder)

70: speech recognition device

In order to achieve the above object, the present invention provides a digital signal by sampling and quantizing a speaker drive signal for driving a speaker of the voice recognition television receiver in a voice recognition television receiver controlled by recognizing a voice command of a viewer. A first A / D converter for converting the data into a second one; The digital signal using N filter coefficients consisting of Noise estimation signal from , A variable coefficient digital delay line filter for generating a signal; The digital signal is sampled and quantized by the viewer's voice command input to the microphone of the television receiver and the interference sound of the speaker sound which is directly or indirectly ringed through the speaker of the television receiver. A second A / D converter for converting the data into a second one; Digital signal of the interference sound And an interference sound estimation signal which is an output signal of the variable coefficient digital delay line filter. Signal of difference An adder / subtractor for calculating And the difference signal At a certain moment to minimize Coefficient of Using Coefficient of moment Remind And It provides a speaker interference noise removing device comprising a filter coefficient generator calculated from and provided to the variable coefficient digital delay line filter.

Hereinafter, with reference to the accompanying drawings Figures 1 to 5 will be described in detail a preferred embodiment of the speaker interference noise canceling apparatus and method according to the present invention.

1 shows an embodiment of a voice controlled television receiver equipped with a built-in or external microphone according to the present invention. Referring to FIG. 1, an external microphone 10 or an internal microphone 20 is mounted to detect a voice command of a viewer, that is, a command such as power on / off, channel change, volume adjustment, and screen adjustment.

On the other hand, as described above, the speaker sound from the left speaker 30 and the speaker 31 and the indirect speaker sound reflected from the room are combined with the voice command of the viewer and input to the microphones 10 and 20. At this time, the television receiver 32 according to the present invention is characterized in that it is provided with a speaker interference canceling device capable of extracting only a viewer's voice command from the interference sound from which the speaker sound is synthesized.

Interference signals input to the microphones 10 and 20 may be regarded as superimposed by signals generated by the attenuation, delay, and phase change of the original signal driving the speaker, and in the absence of a voice command Speaker driving signal In other words, the self-sound interference signal input from the microphones 10 and 20 Is

It is expressed as here Represents the attenuation or phase change along the propagation path of the sound, Represents the delay time.

Fig. 2 is a block diagram showing the components for removing the speaker interference sound of the voice controlled television receiver according to the present invention. Speaker interference cancellation device 60 according to the present invention is an electrical signal for driving the speaker (31, 32) Extract this separately By precisely estimating and removing the estimated signal from the microphone input signal, it eliminates its own interference.

Since the voice command signal 51 is not related to the speaker driving signal 41, only the pure voice command signal 52 remains in the signal passing through the speaker interference canceling device 60 even when the viewer's voice signal is input. do. As a result, the voice command signal 52 from which the speaker interference sound is removed is transmitted to the voice recognition device 70 to increase the command recognition rate.

The speech recognition apparatus according to the present invention may use hardware implemented in other conventional applications or software implemented in a microprocessor. Finally, the voice command from which the interference sound is removed is converted into data required for television control by the voice recognition device 70.

3 is a diagram showing the configuration of a speaker interference canceling device according to the present invention. Referring to Figure 3, the speaker drive signal Is then scaled to an appropriate size in amplifier 40 and then sampled and quantized by analog-to-digital converter (A / D converter) 42 to obtain a digital signal. Is converted to.

here, silver Second sampled digital value. Then, the digital signal converted by the A / D converter 42 Is an interference sound estimation signal by an adaptive digital tapped-delay line filter 62; Is calculated. That is, the variable coefficient digital delay line filter 62 according to the present invention is an input digital signal. Output digital signal from Generates.

At this time, Etc. are called filter coefficients. By adjusting these N filter coefficients, It adjusts so that it may become the signal which estimated the interference sound by this speaker.

On the other hand, according to the present invention, a precise interference sound estimation signal The filter coefficient values necessary to generate Δ are calculated by the filter coefficient generator 61, and a detailed configuration thereof will be described later with reference to FIG.

The variable coefficient digital delay line filter 62 according to the present invention may be implemented by a digital arithmetic circuit such as a multiplier and an adder or a microprocessor program. Meanwhile, the interference signal input to the microphone Is adjusted to an appropriate intensity through the amplifier 64 in the same manner as the speaker driving circuit described above, and then sampled and quantized by an analog-to-digital converter (A / D converter) 65 to obtain a digital signal. Is converted to.

As shown in Equation 1, the interference sound signal is a signal for driving a speaker. Since signals generated through attenuation, delay, and phase change are superimposed, the interference sound estimation digital signal obtained through the above-described process in the adder / adder 63 is added. Digital signal If you remove it from, the sound itself is removed.

As a result, only a voice signal without interference is present at the time of voice command input. Signal with this self-interference removed Is input to the speech recognition device 70 at the next stage, and is also input to the filter coefficient generator 61 at the same time, and the interference sound estimation signal coinciding with the self-interference sound To cause this And filter coefficients.

4 illustrates a variable coefficient digital delay line filter according to the present invention. 4, the variable coefficient digital delay line filter 62 according to the present invention is a speaker drive digital signal. And filter coefficients provided by the filter coefficient generator Interference Sound Estimation Signal Using Indicates that it can be implemented using a multiplier and an adder.

5 is a diagram illustrating an embodiment of a filter coefficient generator according to the present invention. 5, the filter coefficient generator according to the present invention is an interference sound digital signal And estimated digital signal Difference between The filter coefficients can be gradually adjusted to minimize the square or cumulative square of.

As a preferred embodiment according to the present invention, a least mean square (LMS) method or a recursive least square (RLS) method may be used as an error minimization method for generating filter coefficients. More preferably, the least mean square (LMS) method can be used, at which point in time Coefficients Next moment using New coefficient in silver And error Using

only, The filter coefficient is newly determined as follows. here, Is a variable that controls the degree to which the filter coefficients change at once. Usually, a very small random value can be used.

Meanwhile, an initial operation step for removing speaker interference sound according to the present invention, namely Can start with all filter coefficients set to zero. The new filter coefficient thus obtained is transferred to the variable coefficient digital delay line filter 62 shown in FIG. Will be generated.

If the process for calculating the aforementioned speaker interference sound estimation signal is repeated, an error The absolute value of becomes smaller and smaller, and the coefficient correction by Equation 3 is finely stabilized. At this time, the interference sound digital signal And estimated interference Is matched so that when the viewer May be a voice command signal with its own sound removed.

Finally, the voice command signal 52 from which the self-interference sound is removed is transmitted to the voice recognition device 70 to be converted into data necessary for television control. As a preferred embodiment according to the present invention, the speech recognition apparatus may use hardware implemented in the prior art or software implemented in the microprocessor.

When the voice command recognition operation is performed, the result is transferred to a general purpose microprocessor that controls the television receiver to perform basic operations such as powering on / off, changing channels, adjusting the volume, and adjusting the screen of the television receiver. The viewer can turn on or off the entire function by using the above-described voice control function of the voice control television set in the existing remote controller or the main body.

The foregoing has outlined rather broadly the features and technical advantages of the present invention to better understand the claims of the invention which will be described later. Additional features and advantages that make up the claims of the present invention will be described below. It should be appreciated by those skilled in the art that the conception and specific embodiments of the invention disclosed may be readily used as a basis for designing or modifying other structures for carrying out similar purposes to the invention.

In addition, the inventive concepts and embodiments disclosed herein may be used by those skilled in the art as a basis for modifying or designing other structures for carrying out the same purposes of the present invention. In addition, such modifications or altered equivalent structures by those skilled in the art may be variously changed, substituted, and changed without departing from the spirit or scope of the invention described in the claims.

As described above, the present invention generates an electrical signal that mimics the speaker interference sound by using an internal electrical signal for driving the speaker of the television receiver, and removes it from the input signal converted into electricity by the microphone, thereby interfering with the speaker generated sound. The removed voice command signal is obtained. As a result, voice commands with the speaker interference removed are delivered to a voice command recognition module that can be implemented in hardware or software to control basic operations such as powering on / off, changing channels, adjusting the volume, and adjusting the screen of the television set. It generates information.

Claims (11)

In a voice recognition television receiver controlled by recognizing a voice command of a viewer, A digital signal is sampled and quantized by a speaker driving signal for driving a speaker of the speech recognition television receiver. A first A / D converter for converting the data into a second one; Consist of The digital signal using two filter coefficients Noise estimation signal from A variable coefficient digital delay line filter for generating a signal; The digital signal is sampled and quantized by the viewer's voice command input to the microphone of the television receiver and the interference sound of the speaker sound which is directly or indirectly ringed through the speaker of the television receiver. A second A / D converter for converting the data into a second one; Digital signal of the interference sound And an interference sound estimation signal which is an output signal of the variable coefficient digital delay line filter. Signal of difference An adder / subtractor for calculating And The difference signal At a certain moment to minimize Coefficient of Using Coefficient of moment Remind And A filter coefficient generator calculated from and provided to the variable coefficient digital delay line filter Speaker interference cancellation device comprising a. The method of claim 1, wherein the filter coefficient generator is the difference signal using a least mean square (LMS) method Apparatus for removing speaker interference noise, characterized in that to minimize the. The method of claim 1, wherein the filter coefficient generator is the difference signal using an iterative least squares (RLS) method Apparatus for removing speaker interference noise, characterized in that to minimize the. The filter coefficient generator of claim 1, wherein the filter coefficient generator For calculating the filter coefficient of the moment , Calculate the new filter coefficient from the relation of A loudspeaker interference canceller, characterized in that the coefficient of the moment all start with a zero. The apparatus of claim 1, wherein the variable coefficient digital delay line filter is implemented by a digital arithmetic circuit or a microprocessor program including a plurality of multipliers and adders. The apparatus of claim 1, wherein the filter coefficient generator is implemented as a digital arithmetic circuit or a microprocessor program including a plurality of multipliers, adders, and delays. The apparatus of claim 1, wherein the voice command of the viewer comprises any one or a combination of operation commands of power on / off, channel change, volume adjustment, and screen adjustment. The apparatus of claim 1, wherein the first A / D converter or the second A / D converter further includes an amplifier configured to adjust the strength of an input analog signal at a front end thereof. The apparatus of claim 1, wherein the microphone is mounted inside, outside, or a remote controller of the television receiver main body. A method of extracting and separating only a viewer's voice command signal from an interference sound interrupted by direct and indirect sounds of a speaker sound generated by a speaker of a voice recognition television receiver controlled by recognizing a viewer's voice command, (a) sampling and quantizing speaker driving signals for driving the speakers of the speech recognition television receiver; Converting to; (b) digital signal by sampling and quantizing the interference command of the viewer's voice command input to the microphone of the television receiver and the speaker sound which is ringed directly or indirectly through the speaker of the television receiver; Converting to; (c) Composed of Using two filter coefficients Estimate the interference from the signal of Generating from the equation; (d) the estimated signal And the digital signal Signal of difference To yield the above To minimize this Instantaneous filter coefficients of Instantaneous filter coefficients and Calculating a new from; (e) repeating step (c) and step (d) Repeat until is minimized Speaker interference removing method comprising a. The method of claim 10, wherein step (d) Instantaneous filter coefficients silver A speaker interference noise removing method, characterized in that calculated from the relational expression of