JPS6259927B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an audio remote control device for a television receiver, etc., and in particular, to provide an audio remote control device that can perform stepwise control of analog quantities and display thereof using voice commands. The purpose is to

Commander (or operator, hereinafter referred to as a person who remotely controls a controlled device (for example, a television receiver, etc.) using real voice as source information. ) by sampling and quantizing the command (or instruction) content (voice signal) and storing it in advance as a standard (digital signal) pattern, and then sampling the voice instruction content (voice signal) that will be issued later. After that, quantize it, adjust the time axis as necessary, and digitize it.
It has been proposed to temporarily store the pattern in a memory such as RAM, and perform on/off control when it is compared with the standard pattern and matches within a certain tolerance.

Examples of such voice recognition means include an input unit 1 including an acoustic-electrical signal converter (for example, a microphone) that converts input voice into an electric signal, and a feature extraction unit that extracts the characteristics of the voice signal, as shown in FIG. unit 2, standard pattern storage unit 3 that stores standard patterns of voice features registered in advance; recognition processing unit 4 that compares the feature pattern extracted from the input voice with the standard pattern and identifies the input voice; recognition result; Based on this, the main component is an output control section 5 that controls the power, channel, volume, etc. of a television receiver, and an input signal amplitude normalization circuit 6 and a time axis adjustment section 7 to improve the recognition rate. , a registration control section 8 is added for registering standard patterns of audio features in advance.

Many methods can be considered as parameters for extracting voice features, such as frequency spectrum distribution, correlation function, number of zero crossings, formant frequency, or linear prediction coefficient. Among these methods, the frequency spectrum of voice is separated using multiple frequency filters. The so-called filterpunk method, which extracts a pattern and examines its correlation with a standard pattern, is often used as a method that can obtain a high recognition rate with a relatively simple configuration. The control contents of such a voice control device include turning the power on and off, specifying a change of channel number,
For example, the power supply can be controlled by saying things like "Dengen Tori (Kiri)", but since the volume is originally an analog continuously variable control, no suitable control method using voice has been proposed. .

According to the present invention, such an analog quantity is numerically indicated by voice, the contents are discriminated by a voice recognition device, and extracted as, for example, a decimal value, and a plurality of variable attenuators or variable gain amplifiers are used. By changing the combination of the above, the level of the analog quantity is controlled and the level is displayed.

Hereinafter, details of the present invention will be explained with reference to FIGS. 2 and 3 showing main circuit block diagrams.

In this embodiment, a voice control device for a television receiver incorporating a voice recognition device that employs a filter bank method (frequency spectrum method) as a feature parameter for voice recognition will be taken as an example. If the controlled quantity is an analog quantity, the controlled device is not selected.

The input section 1 , which is usually attached to the front of the device, has two microphones 10, one directional and one omnidirectional.
and 11 as shown in the figure, and an amplifier 12. That is, the omnidirectional microphone 11 is connected in opposite phase to the directional microphone 10, and therefore, audio signals from outside the directional characteristic range, that is, signals other than control command voices, are canceled out, and signals within the directional characteristic range are canceled out. The signal-to-noise ratio of the control command voice is thereby increased. At that time, in order to prevent malfunctions caused by similar command words in the audio played from the TV receiver's speakers, the important word (most frequent word) "Dengen" is used.
Alternatively, when commands such as "power,""channel,""only," or "volume" are identified (note: large tolerances have been set for these terms), the A so-called initial muting circuit is provided as shown in FIG. 3 as a circuit diagram of a main part, and the speaker output is temporarily cut off or greatly attenuated. This point will be explained in detail later. In addition to the amplitude normalization function, the feature extraction unit 2 includes a plurality of filters 13-1, 13-
2,..., 13-N and a level detection circuit 14 that detects the total amplitude of the input signal, an A-D (analog-digital) converter 15 that converts the output of each filter into a digital signal, and the A-D converter. an amplitude normalization circuit 16 configured of an analog divider or the like, which is arranged in front of the filter output amplitude and normalizes the filter output amplitude by taking the ratio between the output of each of the filters and the output of the level detection circuit 14; The multiplexer 17 is inserted between the amplitude normalization circuit and the filter group and switches the connection of each output of the filter. With such a configuration, each filter component of the audio signal input from the input section 1 is sequentially sampled at appropriate time intervals (often around 10 milliseconds), and further quantized by quantizing each sampling value. After being converted into digital code,
It is stored in a storage memory 19 (usually RAM: random access memory) via an I/O port (not shown) of a microcomputer or central processing unit (CPU) 18.

In the process of A-D conversion, when each sampled quantity is quantized, it is possible to uniformly quantize it, but when a separate manual adjustment means is provided, control instructions for that adjuster are given in stages. Quantization can also be measured non-linearly depending on the relationship between the quantity (for example, the rotation angle of the volume) and the control level.

Another standard pattern memory 3 is connected to the CPU (Central Processing Unit) 18, and the voice commands (control commands) of the commandant are stored in advance in a sampled and quantized form together with a code specifying the control contents. The control command voice (voice command) is registered in the standard pattern memory in the case of controlling a television receiver, for example, as follows. Figure 4 shows an example of the control panel of a television receiver. , an input microphone 20, a registration mode switch 21, a dispatcher (speaker) number designation switch 1 and 2 for selecting the dispatcher.
2-1 2, 22-2... Control command designation switch "Power" designation switch 2 corresponding to power on/off switching, volume change, channel switching.
3 "Volume" designation switch 24 "Channel" designation switch 25 as well as numeric buttons 26-1, 26-2, 26- for specifying volume and channel
3...26-11, 26-12 are arranged together with respective corresponding indicator lamps 27-1, 27-2,...27-12. Also, the "OK" indicator lamp 28 at the bottom
upon successful completion of recognition or registration;
Similarly, the "REPEAT" indicator lamp 29 lights up to indicate when there is a defect. To register a standard pattern using the registration control unit 30 , first press the registration switch 21 to enter the registration mode, then specify the speaker number with the specification switch 22-1 or 22-2, and then press the registration switch 21 to enter the registration mode. , then press the "power" switch 23 in order to select, for example, "DENGE" or "POWER", and press the "volume" switch 24 to select "Only" or "Volume".
When the "channel" switch 25 is pressed, the registration control circuit 8 in FIG.
and outputs the channel pushbutton switch circuit 31.
The output is switched to the registration control circuit side via the switching circuit 32. As a result, the numerical designation buttons 26-1, 26-2 included in the switch circuit 31...
(Fig. 4) When you press ``1'', ``ni'', etc., each sound goes through the input section 1, the feature extraction section 2, and the control contents (power, volume channels 1, 2, etc.).
3,...) are stored in the standard pattern memory 3 together with the corresponding codes.

Now, in the normal recognition mode, the aforementioned control voice is input, and the feature extraction filters 13-1, 13-2...
The signal string extracted and digitized by 13-N is stored in a storage memory 19 such as RAM, and then
The CPU 18 specifies the input voice by calculating the difference between this stored pattern and the standard pattern for all standard patterns and determining the standard pattern with the smallest difference. In general, human speech is not always the same over time even when uttered in the same language, so it is well known that some kind of time axis adjustment circuit as shown in FIG. 1 must be added. In FIG. 2, the time axis adjustment circuit is omitted for convenience of explanation.

In the recognition mode, voice is always captured, and when the input voice is interrupted, that is, during a pause period, the above-mentioned recognition calculation is executed, and the previous input voice is identified by pattern matching.
At this time, when it became possible to identify the input audio,
That is, when the input audio matches some standard pattern within an allowable error range, the CPU 18 outputs an instruction to the output control circuit 33 to control each corresponding control element of the television receiver. For example, when the input voice "Dengen Iri (Kiri)" is recognized, the output control circuit 33 controls the power supply circuit 34 of the television receiver to turn on and off. Also, “Channel・
※※'' (※※ is a number from 1 to 12) When the input voice is recognized, the output control circuit 33 outputs it to the channel switching circuit 35, and thereby the tuner 3
6 is switched and controlled.

By the way, in order to adjust the volume, the controlled audio device of the apparatus of the present invention must be equipped in advance with a visible display that numerically displays the volume. The appropriate number of steps to replace the continuously variable volume with a stepwise numerical display (ie, digital display) would be around 10. As a concrete display device, for example, a 7-segment numerical display element, or a one-dimensional array of light emitting elements (LEDs, etc.) of the same number as the number of steps can be considered. Here, the case of a television receiver is taken as an example, and the channel display lamp 27 shown in FIG.
-1, 27-2...27-12 is also used as a volume display.

Number display LED27-1, 27- in the figure
2, . . . , 27-12 are shown as channel indicator lamps in FIG.
7-2, or 37-12, OR gate 38-1,
38-2..., or 38-12 and resistor 39-1,
39-2,... or any 1 after 39-12
Two LEDs are lit, indicating the channel. That is, the volume level display control signal b (details will be described later) of the output control circuit is normally at a digital "0" level, and therefore each AND gate 37-1,
37-2...Controls 37-12. Inverter 4
The output of 0 is a digital "1" level. Next, as described above, when the voice recognition device recognizes a control command voice signal corresponding to, for example, "volume", the output control circuit 33 outputs the volume level display control signal b for a certain period of time (several seconds) to the digital "1" level ( In the future H
Gates 37-1, 37-2
..., and at the same time as closing 37-12, open AND gates 41-1, 41-2 and 41-12,
Based on the output of the B-D (binary-decimal) conversion circuit 43 of the Q output of the D-FF (D flip-flop circuit) 42-1, 42-2..., 42-4, which is a latch circuit, the numerical display LED 27- 1 or 27-
2,... or 27-12. As described later, the latch circuits 42-1, 42-
2,...42-4 sets the current volume level to 2.
It is maintained in decimal format. Therefore, LED27-1,
Of course, 27-2,... or 27-12 will numerically display the volume level by the number of lights lit.
Needless to say, segment display may be used. The voice recognition device continues to lower the volume level to 1.
When recognizing the command voice in stages 2 to 2, the output control circuit 33 outputs the volume level latch control signal C, which is normally a digital "0" level (hereinafter abbreviated as L level), for a short period of time (D-FF latches). (sufficient time for
After 4-1, 44-2..., 44-4, the above D-
Clock FF42-1, 44-2..., 42-4. At the same time, the output control circuit 33
The volume level compared and recognized is output as a binary code as volume level signals d ₁ , d ₂ ,..., d ₄ and opened by the volume level latch control signal mentioned above.
AND gates 45-1, 45-2,..., 45-
4, and OR gates 46-1, 46-2,...,
46-4 and inverters 47-1, 47-2, . . . 4 to ensure that the latch circuit 42 or 50 latches the falling edge of the clock as a D input.
7-4, 48-1, 48-2,..., 48-4 then D-FF42-1, 42-2,... or 42-4
Enter the D input. Volume level signal d ₁ , d ₂ ,...,
Latch circuit (D-FF) that maintains the output state of _d4
By the Q output of 42-1, 42-2, ... or 42-4, the volume level of the voice recognition as described above is displayed numerically on LED 27-1, 27-2, ... or 27-.
12 is lit and displayed. Similarly, the volume level latch control signal C is clock inputted to the D-FF 50, which is a state holding circuit, via the OR gate 49. At this time, if the output of the OR gate 51 is at the L level, the output of the one shot circuit 52 is also at the L level. level, the D input terminal passing through inverters 47-5 and 48-5 is at L level, and therefore the D input terminal is at L level.
-The Q output of FF is 0 (=1). As a result, the analog switch 53 is closed and the output of the D-FF50 is 1, so the AND gates 54-1, 54-
2, . . . 54-12 is the output of the binary conversion circuit 43. Depending on the volume level, the analog switch 55 is
-1, 55-2,... or 55-12 opens. On the other hand, the output of the audio FM demodulation circuit 56 of the television receiver passes through the first-stage audio amplification circuit 57, the variable resistor 58, and the variable terminal output is further controlled by the volume adjustment volume 59 or resistors 60-1, 60-2, . . .
Voltage divider circuit (or attenuation circuit or variable gain circuit is also possible) consisting of 60-11, analog switch 5
3 or analog switch 55-1, 55-2,...
55-12, and is amplified by an audio output amplifier 61 and outputted from an audio output device 62 such as a speaker, the analog switches 55-1, 55-2, etc. ,55-
12, an appropriate volume level divided by resistors 60-1, 60-2, . . . , 60-11 is determined. The volume level control signal b returns to the L level after a few seconds (after a period of time during which the commander can confirm the completion of control),
Accordingly, numerical display LED27-1, 27-2,
..., 27-12 returns to the channel display from the volume level display, but the volume level is determined by the latch circuit 42.
The levels corresponding to the output states of -1, 42-2, . . . 42-4 are maintained. The variable resistor 58 determines the variable range of the volume. Next, when the volume adjustment volume 59 is manually operated, the change in the output of the variable resistor 63 connected between the constant voltage power supply V _R and the ground is controlled by the resistor 64, the capacitor 66, and the resistor. This is detected by a comparator 66 whose both ends are connected to both input terminals as shown, and a subsequent manual operation detection circuit 67 outputs a manual operation signal e at H level while the volume adjustment operation continues. When the signal rises, the one-shot circuit 52 operates via the OR gate 51 and generates a pulse, which is held by the latch circuit 50 described above and becomes Q=1 (=0), opening the analog switch 53 and adjusting the volume. The audio output amplifier 61 drives the speaker 62 at the volume level determined by the adjustment volume 59. AND gates 54-1, 54- due to =0 output of D-FF50
It will be clear that analog switches 55-1, 55-2, . . . , 55-12 are all closed. According to the manual operation signal e, the AND gate 6 is activated as shown in the figure.
8-1, 68-2...68-4 are opened and the volume level (binary code) which is the output of the A- _D converter 69 biased to a constant voltage VR is set to the latch circuit 42 at the falling edge of the e signal. -1,42-2,...,42-4
It is stored in memory. As a result, the volume level adjusted and determined by the volume adjustment volume is digitized and held and stored in the latch circuit. Said OR
One of the inputs of the gate 51 is connected to the power-on signal f output by the power supply circuit 34 of the television receiver, and a pulse signal f generated for a certain period of time after the power is turned on is used to maintain a latch at the same volume level and turn on the analog switch. 53 determines the gain of the audio amplification circuit.

Next, details of the above-mentioned initial muting circuit will be explained with reference to FIG. In addition, the third
In the figure, parts common to those in FIG. 2 are given the same reference numerals (figure numbers), and their explanations will be omitted.

As mentioned above, in the voice command recognition mode, the voice signal is constantly captured while the command continues, and the recognition calculation is executed by the CPU 13 when the input voice is interrupted, that is, at intervals of a certain command unit (pause period). The input voice commands up to that point are identified by the pattern matching method.

As mentioned above, in order to avoid malfunctions due to similar sounds emitted by persons other than the commanding person, such as the speaker output of the TV receiver that is the controlled device, the comparative tolerance of pattern matching is set somewhat large for the most frequently used command word.

At this time, when the input voice can be specified, that is, when the input voice matches some standard pattern within an allowable error range, the CPU 20
controls the output control circuit 33 to mute the audio output of the television receiver for a certain period of time. In the case of FIG. 3, the output control circuit 33 lowers the bias of the output amplification transistor 57 of the audio demodulation amplification circuit 56 of the television receiver, thereby increasing the output power of the speaker 59 connected from the collector of the transistor via the capacitor 58. Stop audio output. Note that an earphone 60 connected to the output side of the audio circuit 56
Muting is not necessary for the circuit.
Normally, control command language is made up of a series of multiple words, such as "dengen"/"iri", "dengen"/"kiri", "channel"/"ichi", "channel"/"ni", etc. Therefore, for example, when the input voice ``channel'' is recognized, the speaker voice is muted, and the subsequent voice input of ``1'' or ``2'' eliminates the sound generated by the TV receiver, so the SN ratio is extremely improved and the recognition rate is is greatly improved. This muting of audio is not a functional drawback, regardless of the control contents, such as changing the power on/off channel or changing the volume. In the specific example shown in Figure 3, the volume can be changed by specifying 12 levels like the channel, such as "only" and "sun" (volume 3) (therefore, the channel display device can also be temporarily used as a volume display device). ) is also possible. Based on the result of input voice recognition, the output control circuit 33 controls the power supply circuit 34, channel switching circuit 35, or audio circuit 56 of the television receiver.
Needless to say, control outputs are provided for each.

When the embodiment shown in FIG. 3 is used in combination, the output control circuit of the voice recognition device acts on the voice output circuit of the voice device such as a television receiver equipped with the voice recognition device, and when the input voice is detected, the output of the voice device is outputted. By attenuating the audio to an appropriate level, subsequent input audio can be improved.
This increases the SN ratio and is therefore extremely effective in improving the recognition rate of this type of speech recognition device.

According to the present invention, analog quantities can be controlled in stages by voice commands, and the display thereof can also be visually displayed digitally. Furthermore, even when switching to manual control, automatic step-by-step control and display is possible.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the main part of the speech recognition device, Fig. 2 is an example of the main part implementation circuit of the present invention, Fig. 3 is an example of the initial mutating circuit, and Fig. 4 is the operation panel of the controlled device. Represents a front view. 1 ...Input section, 2 ...Feature extraction section, 3 ...Standard pattern memory, 18...CPU, 37, 41, 45, 5
4,68...AND gate, 44,46...OR gate, 42,50...latch circuit, 53...analog switch, 60...voltage dividing resistor, 43,69...A-D conversion circuit.

Claims (1)

[Claims] 1. An input unit that inputs a channel command voice and a volume command voice that commands the volume step by step, and converts the input voice into an electrical signal, a feature extraction unit that extracts the characteristics of the electrical signal, and a standard pattern storage unit that stores voice characteristics as a standard pattern; a recognition processing unit that compares the characteristic pattern extracted by the feature extraction unit with the standard pattern and identifies the channel command voice and the volume command voice; Based on the recognition result of the recognition processing unit,
In a control device comprising a channel and an output control section for controlling volume, the binary value output for volume control of the output control section is decimated and the volume is controlled stepwise by the decimal evolution signal, and the volume is 1. A voice control device for a television receiver or the like, characterized in that an input circuit of a channel display means is switched according to a recognized output of a command voice, and a volume is displayed using the channel display means according to the decimal evolution signal.