JP2003131682A - Equipment control system using voice command - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an equipment control system using, specifically, voice commands as to a voice recognition system. SOLUTION: The equipment control system has a voice recognition and collation part 7 which is supplied with the conversion output of an A/D conversion part 2 corresponding to a specific command for vocal equipment control, collates it with data of a command file part 6 stored with the codes of the equipment control corresponding to respective commands for the equipment control, outputs an equipment control code corresponding to a command for the equipment control obtained by recognizing a digitized signal as a voice and controls object equipment with a voice. A command file part and the voice recognition and collation processing part are provided on a communication line, and the equipment control code corresponding to the command for the equipment control obtained by supplying the A/D conversion part output to the voice recognition and collation part, collating it with the command file part and performing the voice recognition processing is outputted to the object equipment.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition system, and more particularly to a device control system using voice commands for controlling a target device by voice.

[0002]

2. Description of the Related Art With the recent development of voice recognition technology and peripheral technology, it has been attempted to operate equipment of general electronic equipment such as consumer equipment and industrial equipment by a voice command. M
When a voice recognition system capable of controlling a device by a voice input command is mounted on a device having an MI (man-machine interface) (for example, a consumer stereo), the device is equipped with a voice input device, a digitizing device, and It incorporates and executes all recognition processing devices.

FIG. 4 shows an example of the configuration of a conventional device control system (speech recognition system) for controlling a device by voice. A command input 31 by a predetermined voice by the device user is transmitted by the voice input unit 32
Is supplied to. The voice command signal input to the voice input unit 32 is converted into a digital signal by the A / D conversion unit 33 with a specific sampling frequency and an amplitude bit, and is supplied to the voice recognition engine 34.

In the voice recognition engine 34, the command word file specified by the device data obtained from the input data is acquired from the voice recognition data file 35, and the data of the specified command word file is stored.
The data obtained by voice recognition processing of the input voice command signal that has been A / D converted, that is, the character command data corresponding to the converted voice command word is collated. Then, the command word that best matches is output to the device as a device control code unique to the device controlling the device.

[0005]

Since the conventional voice recognition system is built in each device, it depends on the device.
The ability varies depending on the price, etc., and the voice recognition rate is generally low. Further, in order to increase the recognition rate of the voice recognition processing engine 34 and the recognition vocabulary of the voice recognition data file 35, the system becomes more complicated and the material cost also rises. Especially in general consumer equipment, since the voice recognition ability that can be mounted is naturally limited in terms of material cost, a low voice recognition rate and a small vocabulary correspondence are problems.

Therefore, the present invention has been made to solve the above problems, and in particular, the command file unit 6 and the voice recognition / verification unit 7 are provided on the communication line, and the voice recognition / verification unit 7 is provided. The digitized signal output from the A / D conversion unit is supplied, collated with the command file unit 6, and a device control code corresponding to the device control command obtained by voice recognition processing of the digitized signal is obtained. It is possible to provide a device control system by voice commands output to the target device, simplify part of the device side configuration by using a part of the configuration of the device side commonly in external arrangement, and control the device by voice. The purpose is to

[0007]

In order to solve the above-mentioned problems, according to the invention described in claim 1, a predetermined command for device control by voice is supplied and A / D converted into a digitized signal. The A / D converter 2, the device control command corresponding to the digitized signal obtained by voice recognition processing of the digitized signal, and the device control codes corresponding to the device control commands It has a voice recognition / verification unit 7 that collates the stored data of the command file unit 6 and outputs a device control code corresponding to the device control command, and controls the target device by a voice command. In the device control system using a voice command, the command file unit 6
Also, the voice recognition / collation unit 7 is provided on the communication line, and the digitized signal output from the A / D conversion unit transmitted to the voice recognition / collation unit 7 is supplied to recognize the digitized signal by voice. The device control command corresponding to the digitized signal obtained by processing is collated with the data of the command file unit 6, and a device control code corresponding to the device control command is output to the target device. A device control system based on a voice command is provided, wherein the invention described in claim 2 outputs the corresponding device control code in the device control system based on the voice command according to claim 1. The corresponding device control code or the command corresponding to the device control code is displayed to the device user by voice or character display. There is provided a device control system according to the voice commands, characterized in that the inner.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION The preferred embodiments of the device control system according to the present invention will be described below with reference to the drawings. A specific system configuration of the first embodiment of the device control system by the voice command of the present invention will be described below with reference to FIG.

The first embodiment of the device control system according to the present invention shown in FIG. 1 is an input voice processing unit A, a voice recognition processing site unit B, and an input voice processing unit A and a voice recognition processing site unit. It is composed of the Internet, which is a communication line connecting B with B / home LAN (Local Area Network).

The input voice processing unit A of the device control system according to the voice command of this embodiment includes a voice input unit 1, a voice A / D conversion unit 2, a voice data IP packetization processing unit 3, and an Internet / home LAN interface 4.
Composed of and. The voice recognition processing site section B of the voice command equipment control system of this embodiment is composed of an internet connection interface 5, a corresponding equipment control command file (library) 6 and a voice recognition / verification section 7.

Furthermore, the device control system by the voice command of this embodiment is the Internet / home L
The input voice processing unit A and the voice recognition processing site unit B are connected by the AN, and this Internet / home LAN
Signals are exchanged between the input voice processing unit A and the voice recognition processing site unit B via the.

The input voice processing unit A of the device control system by voice command is connected to an electronic device (target device) to be controlled, and the voice recognition processing site unit B of the device control system by voice command is the input voice processing unit A. Separated from Internet / Home LA
It is arranged and configured on N.

First, a device user sends a predefined command word (for example, "Pawa-on" corresponding to POWER ON) to a device to be controlled by a device built in the device such as a microphone and a microphone amplifier. The voice is input to the voice input unit 1 of the device control system by voice (command) (voice input a).

The voice signal of the analog command word input by the user of the equipment is subjected to preprocessing such as level adjustment, frequency weighting optimal for voice recognition, band limitation, noise removal, etc. in the voice input section 1 and preprocessed. The output of the audio input unit 1 is supplied to the audio A / D conversion unit 2.

The audio A / D converter 2 converts the signal into a digital signal at a specific sampling frequency and amplitude bit,
The converted output of the audio A / D converter 2 remains uncompressed,
It is supplied to the voice data IP packetization processing unit 3.

In the voice data IP packetization processing unit 3, IP
The packetized data is the Internet / home LAN interface 4 connected to the public communication line.
Is transmitted to the voice recognition site section B together with the device code for specifying the device (signal b).

The voice recognition processing site section B is composed of a voice recognition processing / collation section 7, a corresponding device control command file 6 and an internet connection interface 5. The voice recognition processing site section B receives the input voice processing section A from the input voice processing section A. The internet connection interface 5 receives the digitized voice signal transmitted via the internet / home LAN.

The information received by the Internet connection interface 5 is supplied to the voice recognition processing / collation unit 7 (signal c). As shown in FIG. 3, the voice recognition processing / collation unit 7 includes a voice recognition unit 71, a character data output unit 72, and a collation unit 73.

In the voice recognition processing / collation unit 7, the command word file specified by the device code obtained from the input data is obtained from the corresponding device control command file 6 having the command file of each device. (Signal d).

The voice recognition processing / collation unit 7 is supplied with command word file data corresponding to the device code sent from the input voice processing unit A from the corresponding device control command file 6 and specifying the device.

In the voice recognition processing / collation unit 7, the voice recognition unit 71 performs voice recognition processing on the digitized voice signal transmitted from the input voice processing unit A, that is, restored voice data corresponding to the input voice command. Then, the restored voice data is subjected to voice / character conversion processing according to a predetermined algorithm, and character command data corresponding to the voice data is obtained by the character data output unit 72.

Next, the collating unit 73 collates the character command data obtained by converting the voice data subjected to the voice recognition processing with the corresponding device control command word file data supplied from the corresponding device control command file 6. .

Then, the collating unit 73 selects one command word that most optimally matches the character command word data. Here, in the corresponding device control command file 6, character commands that may be generated in advance (allowable command words to be described later) are listed as candidates and are listed together with the official command words, so that verification is possible. The degree of coincidence at time can be increased.

Further, the device control code (device control command) corresponding to the selected command word is determined from the command words of the corresponding device control command file 6,
The determined device control code is supplied to the internet connection interface 5 (signal e).

Internet connection interface 5
Outputs (replies) the device control code received from the voice recognition / collation unit 7 to the device (target device) M that issued the control request via the Internet / home LAN.
(Signal f).

The device M receives the device control code for controlling the device M via the Internet / home LAN connection interface 4, and executes the device operation and the device setting corresponding to the received device control code.

An example of the command file used in the device control system by the voice command of the present invention will be described below.

[0028] Manufacturer: XYZC Model name: SM-2001A

Command: Power on Command word: "Pawaon", "Dengen On" Permissible command words: "Awaon", "Waon",
"Engen-on", "Gen-on" Equipment control code: 01

Command: record Command word: "Kiroku", "Record" Allowable command words: "Iroku", "Ecode" Equipment control code: 02

Command: Play Command word: "Saisei", "Purei" Allowable command words: "aisei", "please" Equipment control code: 03

Command: Stop command word: "Tomare", "Stop", "Teshi" Allowable command word: "Omare", "Utopp", "Topp", "Eishi" Device control code : 04

Command: XX Command word: “………”, “………” Permissible command words: “………”, “………” Equipment control code: 0E

As described above, the device control system according to the voice command of this embodiment is capable of performing voice recognition /
The collation unit 7 and the corresponding device control command file 6 are provided separately from the device M by a communication line, and are commonly used in the device control system by the voice command of a large number of devices, and the number thereof is Since it can be reduced, unlike the conventional system, it is possible to increase the material cost etc. until the performance is sufficiently satisfied, and it is also possible to improve it, and its speech recognition rate is the same as the conventional built-in type that is not separated. It can be dramatically increased in comparison.

Further, in the present embodiment, the corresponding device control command file 6 does not need to correspond one control word to one control command.
Supports multiple words (eg, 1 control command STOP
It is possible to deal with multiple words such as "stop", "stop", "stop", etc.), and there is enough room to allocate multiple vocabulary for one action, and the user can use general vocabulary. The degree of freedom to control the device with
You can build a device control system with very user-friendly voice commands.

Further, according to the present embodiment, the manufacturer code,
Since the manufacturer and target device can be specified by the device code, etc., the device is controlled by voice commands for all products that have an internet connection environment.
Is possible.

The advantages of the device control system using the voice command of this embodiment will be summarized below. (A) The command file part and the voice recognition / verification part of the main part are provided on the communication line such as the Internet or a home LAN, and the system configuration does not need to be incorporated into the device, so that the cost can be greatly reduced. (B) The command file unit 6 and the voice recognition / verification unit can realize a recognition rate that does not depend on the price of the embedded device.

(C) Only the voice recognition rate can be improved regardless of the device control system on the user's device side. (D) Since the command file unit 6 and the voice recognition / collation unit are provided on the communication line such as the Internet or a home LAN, the accuracy of each component can be improved, and stable voice recognition can be obtained.

(E) The number of recognized vocabularies in the command file can be improved. (F) A user-friendly system can be constructed. (G) It is possible to support multiple models and manufacturers. (H) When using a server on the Internet,
Standby power at home can be eliminated.

Next, FIG. 2 shows a concrete system configuration of the second embodiment of the device control system by the voice command of the present invention. The device control system based on the voice command of the present embodiment shown in FIG. 2 includes an input voice processing unit A, a voice recognition processing unit B, and an Internet / connecting the input voice processing unit A and the voice recognition processing site unit B. It is composed of a home LAN.

The input voice processing unit A of the device control system according to the voice command of this embodiment includes a voice input unit 1, a voice A / D conversion unit 2, a voice data IP packetization processing unit 3, and an Internet / home LAN interface 4.
It is composed of Further, the voice recognition processing site section B of the voice command equipment control system of the present embodiment comprises an internet connection interface 5, a corresponding equipment control command file 6 and a voice recognition / verification section 7.

Furthermore, the Internet / home LA of the device control system by the voice command of this embodiment
N is configured to connect the input voice processing unit A and the voice recognition processing site unit B.

The second embodiment of the device control system according to the present invention shown in FIG. 2 differs from the first embodiment shown in FIG. 1 in that the structure of the device M is that of the device M0. The command guide part of the device M0, and the command guide part added to the device M will be described below.

In the device M0 of FIG. 2, a code / command conversion unit M1 and a command voice guide unit M2 for guiding a command corresponding to the device control code by voice are provided in the device body M, and L.
A display device for displaying characters such as a CD is provided with a command character display guide portion M3 for displaying and guiding commands corresponding to the device control code in characters. The device main body M0 in FIG. 2 is shown as having both the command voice guide section M2 and the command character display guide section M3, but naturally it may have either one.

The device control code returned to the device M0 side via the input voice processing unit A of the device control system by the voice command of this embodiment is converted into a corresponding command by the code / command conversion unit M1. The command is displayed in text on the display of.

For the character display, for example, a character corresponding to the returned device control code is displayed in the code / command conversion unit M1.
, And the character of the corresponding command is output to the character display guide unit M3 such as LCD each time to guide the display.

Alternatively, the device control code returned to the device M0 is converted into a corresponding command by the code / command conversion unit M1 and the command is guided by voice using the speaker of the command voice guidance unit M2. indicate.

The device user can visually recognize and confirm the command corresponding to the returned device control code by this character display and voice guidance, and the device user can use the device with peace of mind. The control command for the next device can be requested to the device by voice.

When the device M is a sound-related device, the system of the device main body M can be used without particularly having the command voice guide part M2 and the command character display guide part M3. Further, when the device M has a display system, it is not necessary to provide the command character display guide section M3 by utilizing the display system.

In this embodiment, the voice device control system having the code / command conversion unit M1 has been described above. However, depending on the device, the code / command conversion unit M1 for converting a code into a command is omitted. The configuration may be simplified by simplifying only the code corresponding to the command and guiding it in the character display as it is.

[0051]

As described above in detail, according to the device control system by the voice command of the present invention,
An external processing device connected to the communication line is provided with a command file section and a voice recognition / verification section to execute the voice recognition processing and the collation processing, so the command file section and the voice recognition / verification section are not required for each device side. Yes, the system configuration on the device side can be greatly simplified, and each device control system can be constructed at a significantly low cost.

Further, according to the device control system by the voice command of the present invention, by increasing the accuracy of the command file unit and the voice recognition / verification unit of the external processing device, the voice recognition rate is high independent of the level on the device side,
It is possible to build a device control system with voice commands that has a larger number of recognized vocabularies.

According to the invention of claim 2, the reply command to the device side is displayed by characters on the display,
Alternatively, since the voice is guided by a speaker or the like, the device control command corresponding to the device control code can be surely confirmed through audiovisual sense, and the target device can be given a voice instruction of the next device operation with confidence.

[Brief description of drawings]

FIG. 1 is a system configuration diagram showing a specific configuration of a first embodiment of a device control system using a voice command according to the present invention.

FIG. 2 is a system configuration diagram showing a specific configuration of a second example of the device control system by a voice command of the present invention.

FIG. 3 is a diagram showing an embodiment of a voice recognition / collation unit that constitutes a device control system by a voice command of the present invention.

FIG. 4 is a system configuration diagram showing an example of a configuration of a conventional device control system using a voice command.

[Explanation of symbols]

1, 32 voice input unit 2 voice A / D conversion unit (A / D conversion unit) 3 voice data IP packetization processing unit 4 Internet / home LAN (Local Area Network) interface 5 Internet connection Interface 6 Compatible device control command file library (command file section) 7 Voice recognition / collation section 31 Voice command input 33 A / D conversion section 34 Voice recognition engine 35 Voice recognition data file 71 Voice recognition section 72 Character data output section 73 Collation Part A Input voice processing part B Voice recognition processing site part M, M0 (electronic) device (target device) to which the device control system by the voice command of the present invention is connected M1 code / command conversion part M2 command voice guidance part M3 command character Display guidance section

Claims (2)

[Claims] 1. A / D which is supplied with a predetermined command for device control by voice and A / D-converts into a digitized signal.
A conversion unit, a device control command corresponding to the digitized signal obtained by performing voice recognition processing on the digitized signal, and each device control code corresponding to each device control command are stored. A device control system using a voice command that controls a target device by a voice command, which has a voice recognition / collation unit that collates with data in a command file unit and outputs a device control code corresponding to the device control command. In the above, the command file unit and the voice recognition / collation unit are provided on a communication line, and the digitized signal of the output of the A / D conversion unit transmitted to the voice recognition / collation unit is supplied to perform the digitization. The device control command corresponding to the digitized signal obtained by voice recognition processing of the signal; And collating the-yl portion of the data, instrument control system by voice commands and outputs a control code corresponding to the command of the device control to the target device. 2. A device control system using a voice command according to claim 1, wherein a device control code corresponding to the command is output and a device control code corresponding to the command, or a command corresponding to the device control code. A device control system using voice commands, which guides the device user by voice or text display.