JPH0682575A - Voice controller for illumination/clock - Google Patents

Abstract

PURPOSE:To make the control of lighting of illuminator and notification of time possible by input of synthesized voice or switch operation. CONSTITUTION:In the illumination/clock controller 10 for controlling an illuminator 15 or a clock 16 selectively through input of voice or switch operation, the illuminator 15 is lighted for some time upon receiving a specific sound for commanding lighting of the illuminator 15. Upon receiving another sound commanding notification of time, the time announced with a synthesized voice.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice lighting / clock control device capable of selectively controlling lighting and a clock by human voice.

[0002]

2. Description of the Related Art Conventionally, as disclosed in Japanese Patent Application Laid-Open No. 2-30095 “Time voicing type lighting device”, there has been proposed a device in which a lighting device and a timepiece device having a time notification function by voice synthesis are combined. Moreover, the applicant has already applied for Japanese Patent Application Laid-Open No. 2-309398 “Voice-based lighting control device”.

[0003]

However, in the above-mentioned "time voicing type illumination device", it is necessary to extend the hand within the range that can be sensed by the touch sensor, which is inconvenient. Further, in the above-mentioned "lighting control device by voice," the object that can be controlled by voice is limited to lighting.

An object of the present invention is to make it possible to control lighting of a light and time notification by synthetic voice by either voice input or switch operation.

[0005]

According to a first aspect of the present invention, in a lighting / clock control device with a voice capable of selectively controlling lighting and a clock by either voice input or switch operation, an instruction to turn on the lighting is issued. When a specific voice is input, the light turns on for a certain period of time, and when another specific voice that commands the notification of time is input, it has a function to notify the time at that time with a synthetic voice. It is a thing.

According to a second aspect of the present invention, in addition to the first aspect of the present invention, an illuminance detector for detecting the ambient brightness is provided, and voice input is performed only when the ambient illuminance is below a certain value. When the value exceeds a certain value, only the operation by the switch is allowed.

According to a third aspect of the present invention, in the present invention according to the first or second aspect, a neural network is further used as a recognition unit of the input voice, and the input voice of the input voice is input to the neural network. This is to use the temporal change of the average frequency characteristic within a fixed time.

[0008]

According to the first aspect of the present invention, it is possible to control lighting of a light or time notification by a synthetic voice by inputting a voice or operating a switch. That is, the illumination is turned on when a pre-registered voice (for example, "illumination") is input, and is turned off after a certain period of time. When a preregistered voice (for example, “clock”) is input, the time at that time is notified by a synthetic voice. Similar functions can be realized by operating switches instead of voice. It should be noted that voice registration requires the user to register his / her own voice when using the specific speaker-speech recognition technology, but when using the unspecified speaker-speech recognition technology, the voice is preliminarily recorded. Since it is registered in, the user can save the trouble of registering the voice.

Therefore, in the device according to the first aspect of the present invention, when the user registers the voice (eg, "lighting") registered in advance in the morning, when waking up, at night, or when going to the toilet, the light is turned on for a certain period of time. Therefore, the bedside is bright and safe. Also, the positions of switches and remote controls can be easily checked while the lights are on.

When waking up in a pitch-black bedroom, if a pre-registered voice (for example, "clock") is uttered, the time at that time can be heard by a synthetic voice, which is particularly advantageous for people with poor eyesight. .

Since the reaction is transient (illumination:
It only lights up for a certain period of time, clock: Just notify the time once), and the effect of malfunctioning is small.

An apparatus according to a second aspect is an apparatus having a configuration in which an illuminance detecting section for detecting ambient brightness is added to the apparatus according to the first aspect. By providing the illuminance detection unit, voice input can be performed only when the ambient illuminance is below a certain value, and only operation by a switch is possible when the ambient illuminance exceeds a certain value.

Therefore, in the apparatus according to the second aspect, since the voice input is possible only in the dark, the merit of the voice input can be maximized.

In addition, it seems that there is not much opportunity to make a sound in the dark, and the possibility of malfunction is reduced.

According to a third aspect of the present invention, when performing voice recognition in the voice recognition section of the device according to the first or second aspect,
This is a device that uses a neural network and uses as a input the temporal change of the average frequency characteristic of the input voice within a certain time.

Therefore, in the apparatus according to the third aspect, by using the neural network, it is possible to realize an apparatus which can process the voice recognition processing in real time while maintaining a high recognition rate.

However, the description of the "neural network" in the present invention is as follows (1) to (4).

(1) From the structure of the neural network, the neural network and the hierarchical network shown in FIG.
It can be roughly classified into two types of mutual coupling networks shown in (B). The present invention may be configured by using either of both networks, but the hierarchical network is more useful because a simple learning algorithm as described later has been established.

(2) Network Structure The hierarchical network has a hierarchical structure including an input layer, an intermediate layer, and an output layer, as shown in FIG. Each layer is composed of one or more units. The coupling is only forward coupling such as input layer → middle layer → output layer, and there is no coupling within each layer.

(3) Structure of Unit The unit is a model of a brain neuron as shown in FIG. 6 and has a simple structure. Receive input from other units,
The sum is taken and converted according to a certain rule (conversion function), and the result is output. A variable weight that represents the strength of the connection is attached to each of the connections with other units.

(4) Learning (Back Propagation) Learning a network is to bring an actual output close to a target value (desired output). Generally, the conversion function and weight of each unit shown in FIG. Is learned by changing.

As a learning algorithm, for example, Rumelhart, DE, McClelland, JL and the PDP Re
search Group, PARALLEL DISTRIBUTED PROCESSING, the
Backpropagation described in MIT Press, 1986. can be used.

[0023]

FIG. 1 is a schematic diagram showing an example of the present invention, FIG. 2 is a schematic diagram showing another example of the present invention, FIG. 3 is a schematic diagram showing a voice recognition unit using a neural network, and FIG. FIG. 5 is a schematic diagram showing a network, FIG. 5 is a schematic diagram showing a hierarchical neural network, and FIG. 6 is a schematic diagram showing the structure of a unit.

(First Embodiment) (See FIG. 1) As shown in FIG. 1, the lighting / clock control device 10 includes a voice input unit 11, a voice recognition unit 12, a switch 13, a lighting / clock control unit 14, and a lighting unit 15. , Clock 16, voice synthesis / output unit 17
Is configured.

Then, the lighting / clock control device 10 selectively controls the lighting 15 and the clock 16 by either the voice input to the voice input unit 11 or the operation of the switch 13. At this time, the illumination / clock control unit 14 obtains the recognition result of the voice recognition unit 12, and (a) receives a specific voice (for example, “illumination”) for instructing the lighting of the illumination 15 to turn on the illumination 15. When lighting is controlled for a certain period of time, and (b) another specific voice (for example, "clock") that commands the notification of the time is input, the time at that time is notified by the synthetic voice from the voice synthesis / output unit 17. Control.

That is, the lighting / clock control device 10 operates as follows. The voice input from the voice input unit 11 is the voice recognition unit 1.
In 2, the voice is recognized.

According to the result, the lighting 15 is turned on or the time is notified by a synthetic voice.

The function similar to the above can be realized by operating the switch 13 instead of voice input.

Therefore, the lighting / clock control device 10 has the following effects. When you wake up in the morning, at night, or when you go to the bathroom, you can utter a pre-registered voice (for example, "lighting") to turn on the lights for a certain period of time, so the bedside is bright and safe. Also, the positions of switches and remote controls can be easily checked while the lights are on.

When waking up in a pitch-black bedroom, if a pre-registered voice (for example, "clock") is uttered, the time at that time can be heard by a synthetic voice, which is particularly advantageous for people with poor eyesight. .

The reaction is transient (illumination:
It only lights up for a certain period of time, clock: Just notify the time once), and the effect of malfunctioning is small.

(Second Embodiment) (See FIG. 2) The illumination / timepiece control device 20 differs from the device 10 in that an illuminance detector 21 is additionally provided. The illuminance detection unit 21 detects ambient brightness. Then, the illumination / clock control unit 14 obtains the detection result of the illuminance detection unit 21 and enables voice input only when the ambient illuminance is below a certain value, and when it exceeds a certain value, the operation by the switch 13 is performed. Only possible.

That is, the lighting / clock control device 20 operates as follows. The illuminance detection unit 21 detects the illuminance of the surroundings, and voice input is possible only when the illuminance of the surroundings is less than a certain value, and when the illuminance of the surroundings exceeds a certain value, only the operation by the switch 13 is possible.

When voice input is possible, the voice input section 11
The voice recognition unit 12 recognizes what voice is input from the voice.

In accordance with the result, lighting of the lighting 15 or time notification by synthetic voice is performed.

Therefore, the lighting / clock control device 20 has the following effects. Since voice input is possible only in the dark, the advantages of voice input can be maximized.

Further, it seems that there is not much opportunity to make a sound in the dark, and the possibility of malfunction is reduced.

(Third Embodiment) (Refer to FIG. 3) A neural network as shown in FIG. 3 is used as the voice recognition unit 12 of the lighting / clock control devices 10 and 20, and the input voice is averaged within a certain time as its input. It uses the temporal change of the typical frequency characteristic.

As shown in FIG. 3, the voice recognition unit 12 is composed of a bandpass filter 31, an averaging circuit 32, a neural network 33, and a determination circuit 34, which are connected to each other.

The operation of the voice recognition unit 12, especially the unspecified speaker
The details of the voice recognition method are as follows.

For 100 registrants, the learning sounds (“lighting” and “clock”) were sampled at a frequency of 10 kHz.
After sampling at, it is equally divided into four sections on the time axis.

Frame length 25.6 msec, frame period 12.8
Fourier analysis is performed for msec to obtain a series of 16ch power vectors in the band of 100 to 5000Hz.

These vectors are used as inputs to the neural network (input layer 64 units, output layer 2 units), and sufficient learning is performed so that the output unit corresponding to the input voice is activated.

A sequence of power vectors is obtained in the same manner as for any utterance.

This is input to the network learned by and an output is obtained for each output unit.

The decision regarding the input voice is made depending on which of the two obtained output values is larger.

When evaluated by the utterances of 5 persons other than the registrants, all the voices uttered by each speaker 10 times could be recognized.

Therefore, according to the third embodiment, there are the following effects. By using the neural network 33, it is possible to realize a device capable of performing voice recognition processing in real time while maintaining a high recognition rate.

[0049]

As described above, according to the present invention, it is possible to control lighting of a light and time notification by a synthetic voice by either voice input or switch operation.

[Brief description of drawings]

FIG. 1 is a schematic view showing an example of the present invention.

FIG. 2 is a schematic view showing another example of the present invention.

FIG. 3 is a schematic diagram showing a voice recognition unit using a neural network.

FIG. 4 is a schematic diagram showing a neural network.

FIG. 5 is a schematic diagram showing a hierarchical neural network.

FIG. 6 is a schematic view showing a structure of a unit.

[Explanation of symbols]

 10, 20 Lighting / Clock Control Device 11 Voice Input Unit 12 Voice Recognition Unit 13 Switch 14 Lighting / Clock Control Unit 15 Lighting 16 Clock 17 Voice Synthesis / Output Unit 21 Illuminance Detection Unit 31 Bandpass Filter 32 Averaging Circuit 33 Neural Network 34 Judgment circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H05B 37/02 E 8715-3K

Claims (3)

[Claims] 1. In a lighting / clock control device with a voice capable of selectively controlling lighting and a clock by either voice input or switch operation, when a specific voice for instructing lighting of the light is input, only the time when the lighting is present. A lighting / clock control device using a voice, which has a function of turning on a light and inputting another specific voice for instructing notification of time, and notifying the time at that time with a synthetic voice. 2. The apparatus according to claim 1, further comprising an illuminance detection unit for detecting ambient brightness, enabling voice input only when the ambient illuminance is below a certain value, and when the illuminance exceeds a certain value. A lighting / clock control device by voice, which can be operated only by a switch. 3. The apparatus according to claim 1 or 2, wherein a neural network is used as a recognition unit of the input voice, and an average frequency characteristic time within a fixed time of the input voice as an input to the neural network. Lighting and clock control device using voice that is characterized by using dynamic changes.