JP2614552B2 - Voice recognition remote control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a remote controller for remotely controlling electric equipment, and more particularly to a voice recognition remote controller for remotely controlling a plurality of electric equipment by voice input.

[0002]

2. Description of the Related Art A remote control device (hereinafter referred to as a "remote control device") is generally used in order to allow an operator to operate a remote electrical device without moving the same. Also widely used.

As this remote controller, a wireless type is generally used. In a wireless type remote control device, infrared rays are used, and a control signal code is pulse-modulated and transmitted. This pulse modulation method includes a method of distinguishing between a binary value “1” and a binary value “0” depending on the presence or absence of an optical pulse, and the binary values “1” and “0” depending on the width and position of the optical pulse. How to distinguish (pulse position modulation)
and so on. The pulse position modulation method has few malfunctions and is generally used.

In general, transmission data tells a reception circuit of transmission start, a "start marker" for facilitating reception of each signal transmitted next, and a device for preventing interference with other devices. This includes codes such as a "custom code" defined for each transmission, and a serialized "data code" defined for each transmission function. The transmission method includes a method of transmitting a series of transmission data repeatedly during transmission, a method of transmitting a series of transmission data only at the start of transmission, and thereafter transmitting only a “continuous leader code” indicating that transmission is in progress. There is a method.

[0005] On the receiving side, the operation of the target electrical equipment is controlled by the custom code and the data code of the transmitted data. Usually, transmission data supplied to the receiving circuit is often transmitted by pulse-modulating a carrier modulation wave in order to prevent interference with other devices and malfunction.

In such a remote control device for remotely controlling an electric device, one remote control device is required for one electric device, and it is generally impossible to remotely control another electric device with the same remote control device. It is possible. For example, it is not possible to remotely control an air conditioner (air conditioner) using a television remote control device. Also,
The remote control device is provided with a number of switches in accordance with the contents of the control operation, and a device control signal is selected according to a pressed state of the switch and transmitted to an electric device to be operated. In the case of a video tape recorder, etc., there are many operations such as setting of TV receiving station channel, tape running state, etc.Therefore, a large number of switches are required in the remote control device, and the operability is poor. In this case, there is a problem that an incorrect operation is performed by pressing the switch.

[0007] Further, since a remote control device is required for each electric device, it is necessary for the user to always accurately grasp which remote control device corresponds to which electric device, and the operation is troublesome. There was a problem.

A remote control device for eliminating a large number of switches as described above and operating a plurality of electric devices with a single remote control device is disclosed in, for example, Japanese Patent Application Laid-Open No. 2-1.
No. 71098. In this prior art, a remote operation content is instructed by voice input, and a control signal is generated according to the voice recognition result. In this prior art voice recognition remote control device, a rewritable map for converting a voice recognition result into a device control code is provided in order to support a plurality of electrical devices, and the contents of this map are used as an operation target of the electrical device. It has been rewritten according to the device. In this prior art, a conversion code is read from an IC card into a map.

[0009]

In the prior art voice recognition remote control device, it is possible to remotely control a plurality of electric devices with one remote control device. However, in order to control a plurality of electric devices, a code conversion IC card is required for each electric device to be controlled,
Before use, it is necessary to insert the IC card into the remote control device main body in advance according to the electric device to be operated. Therefore, also in this case, it is necessary to know in advance which IC card corresponds to each electric device, and it is necessary to insert the IC card before use.

[0010] Instead of such a configuration in which a plurality of control code conversion table maps are rewritten by replacing an IC card, the following configuration can be considered. That is, a read-only memory is provided inside the remote control device main body, and a code conversion table corresponding to each of the plurality of electric devices is written in the read-only memory. In use, only the corresponding conversion code table is used in this read-only memory by designating the electrical device with a switch. In this configuration, although the trouble of replacing an IC card can be omitted, it is necessary to switch a switch for selecting a conversion table and confirm a switch position for each electric device, resulting in a disadvantage that operability is poor.

An object of the present invention is to eliminate the above-mentioned disadvantages of the conventional remote controller and to provide a remote controller having excellent operability.

[0012]

According to a first aspect of the present invention, there is provided a voice recognition remote control apparatus, wherein a plurality of basic voice recognition results and a control signal for controlling an electric device to be generated corresponding to each of the basic voice recognition results. Device / word correspondence storage means for storing information indicating the operation state of the electric equipment, and control for storing the control target electric equipment corresponding to each of the control signals in association with each other. Signal / device state correspondence storage means. The control signal / equipment state correspondence storage unit further stores a transmission prohibition condition flag, which is formed in advance based on the operation state of the control target electric apparatus and indicates whether the control signal can be transmitted, in association with each of the control target electric apparatuses. .

The voice recognition remote controller according to the first aspect of the present invention further includes a first voice recognition unit for recognizing an input voice and outputting the recognition result, and a recognition result from the first voice recognition unit. Second speech recognition means for searching the device / word correspondence storage means for a corresponding control signal. The searched control signal includes the electric device identification information of the control target and the information indicating the operation state to be operated.

The voice recognition remote controller according to the present invention further includes a transmission prohibition condition flag associated with the corresponding control target electric device from the control signal / device state correspondence storage device in accordance with the control signal retrieved by the second voice recognition device. Search for
When the transmission prohibition condition flag related to the searched electric device to be controlled indicates that transmission is possible, a transmission means for transmitting the searched control signal is provided.

According to a second aspect of the present invention, in the case where there are a plurality of control signals identified by the input voice and the plurality of control signals can be transmitted, the name of the electric device to be controlled is changed by the operator. And stopping the transmission of the control signal.

[0016]

According to the first aspect of the present invention, the control signal to be generated from the device / word correspondence storage means is searched and the name of the corresponding electric device is specified based on the recognition result of the input voice.
A transmission prohibition condition flag indicating whether transmission is possible or not is formed in advance for each control signal according to the operation state of each electric device. The transmission prohibition condition flag is provided corresponding to each control signal. When the transmission prohibition condition flag associated with the searched control signal indicates that transmission is possible, the control signal is transmitted. Thus, even when a plurality of electric devices are remotely controlled, identification of the electric device to be operated can be performed without switching the conversion map between the speech recognition result and the device control signal in accordance with the electric device to be operated. And a control signal can be transmitted.

According to the second aspect of the present invention, a control signal can be transmitted to a plurality of electric devices, and if one electric device cannot be specified, the name of the transmittable electric device is given to the operator. The presentation and transmission of the control signal is stopped.
As a result, a backup system is established when it is not possible to specify the electric device to be controlled.

[0018]

FIG. 2 is a diagram showing the appearance of a voice recognition remote controller according to an embodiment of the present invention. In FIG. 2, the voice recognition remote control device is stored in a storage case 1. On one surface of the storage case 1, a microphone 2 for inputting voice information, a communication processing unit 3 for transmitting a control signal corresponding to the input voice, and an instruction to permit / cancel transmission of the communication processing unit 3. A switch unit 5 for inputting information to be input, and a display unit 4 for displaying input voice recognition results and the like.
And a speaker 6 for presenting confirmation information from the remote control device to the operator. Since the communication processing unit 3 transmits a control signal according to, for example, an infrared ray system, a portion such as an infrared ray generating diode for transmitting the control signal is visible from one surface of the storage case 1.

The switch unit 5 includes a cancel switch 5a for inputting information indicating transmission cancellation, and an OK switch 5b for inputting information indicating transmission permission. This transmission permission / cancellation is determined by the operator according to the information displayed on the display unit 4. The display unit 4 displays information presented by the remote control device to the operator in a visible form. FIG. 2 shows an example of a state where the recognition result “reservation” is displayed when the input speech recognition result is “reservation”. From the speaker 6, the remote control device converts the confirmation information or the attention information into an audio information form by the audio forming unit 60 and presents the information to the operator.

FIG. 1 is a block diagram functionally showing the overall configuration of a voice recognition remote controller according to an embodiment of the present invention. In FIG. 1, a voice recognition remote control device stores, in a table form, a basic voice recognition result (word) and a control signal generated according to each basic voice recognition result (word).
A word correspondence table 11 and a rewritable device state memory 12 for storing information indicating a state of an electric device to be operated
And a voice recognition unit 10 that recognizes a voice input via the microphone 2 and searches the device-word correspondence table 11 for a control signal corresponding to the recognition result.

The speech recognition unit 10 includes a first speech recognition unit 101 for recognizing a speech given from the microphone 2, and a corresponding control based on a recognition result from the first speech recognition unit 10 from a device-word correspondence table 11. Second speech recognition unit 1 for searching for a signal
02. The second voice recognition unit 102 also has a function of initializing the device status information stored in the device status memory 12.

The basic speech recognition result (word) stored in advance in the device-word correspondence table 11 is generated in advance in correspondence with an operation that can be controlled by the speech recognition remote controller. The operator gives remote operation control information by uttering any one of the basic speech recognition results (words). The data storage form of the device-word correspondence table 11 will be described later in detail.

The device status memory 12 stores information indicating the operation status of each electrical device. The device state instruction data is rewritable, and its contents are rewritten according to a control operation by the remote control device. By making the contents of the device state memory rewritable according to the remote operation status, it becomes possible for the remote control device to determine the type of the electrical device to be controlled as its use progresses, and to recognize the voice recognition result and The conversion of the electric device control signal conversion map is equivalently performed.

The voice recognition remote controller further stores each control signal, an operation target electrical device, and transmission inhibition condition data (flag) set in advance corresponding to each control signal in a table form. The contents of the device state memory are changed according to the control signal-device state correspondence table 14 and the control signal given from the voice recognition unit 10 (the second voice recognition unit 102), and the transmission prohibition condition of the control signal-device state correspondence table 14 A device control unit 3 for searching data (flag) and providing a control signal to the communication processing unit 3 when transmission is possible.

The communication processing unit 3 includes a second speech recognition unit 102
Is displayed on the display unit 4, and when the operator inputs transmission permission information via the switch unit 5, the control signal is transmitted.

In this case, the communication processing unit 3 may be configured to transmit a transmittable control signal given from the device control unit 13 without waiting for input of confirmation information from the switch unit 5. The form of the communication system from the communication processing unit 3 may follow the infrared system as described above, or an optical signal using a laser beam may be used. The type of signal to be transmitted, the encoding method, and the like may be of any type, and any control signal may be used as long as it can be confirmed by the electric device to be operated. Next, the configuration and operation of each unit will be described in detail.

FIG. 3 shows the first speech recognition unit 101 shown in FIG.
FIG. 3 is a diagram showing an example of a specific configuration of FIG. In FIG. 3, a first speech recognition unit 101 converts an analog signal input from the microphone 2 into a digital signal, and then converts the analog signal into a time series of acoustic parameters such as a power spectrum. Word standard pattern dictionary 113 for storing time series of acoustic parameters as standard patterns
And a word recognition unit that calculates the similarity between the time series of the sound parameters given from the sound analysis unit 111 and the standard pattern stored in the word standard pattern dictionary 113, and outputs the closest word as a recognition word (recognition result). 1
12 is included. The acoustic analysis unit 111 generates a time series of a short-time spectrum of the input voice (digital voice input). Normally, in the acoustic analysis, after the high-frequency emphasizing process, a short-time spectrum of the voice is obtained every predetermined time (for example, 10 to 20 ms). For the spectral analysis, a band filter group composed of a plurality of band-pass filters or an 8th to 12th-order linear prediction model is usually used. In the case of bandpass filter analysis, a Euclidean distance between feature vectors of a filter output is usually used as a spectrum distance measure for calculating similarity. The first voice recognition unit may be any unit that can recognize the input voice. Any structure that performs word recognition in accordance with the word-speech recognition method of an unspecified speaker may be used. Instead of the method using the word standard pattern shown in FIG. 3, the word speech is recognized using a phoneme standard pattern. Or other word recognition schemes may be used.

FIG. 4 is a diagram showing an example of the stored contents of the device-word correspondence table shown in FIG. In FIG. 4, the device-word correspondence table 11 includes a kana string KR indicating the pronunciation of the control word,
It includes a display data sequence HDR for arranging recognition results to be displayed on the display unit 4 and a control signal sequence SR for arranging control signals to be generated corresponding to the recognition results for each word. The control signal sequence SR includes a device sequence AR that specifies an electric device to be operated. A control signal to be generated corresponding to the recognition result is arranged for each device row AR.

The words arranged in the word string KR are set in advance and are determined according to the electric equipment to be operated. In the device-word correspondence table 11, for example, when the operator desires to make a timer reservation, the uttered word is “yes”. In this case, since the recognition result is indicated by the kana string, a row corresponding to the corresponding row “yokayaku” in the kana string KR of Table 11 is selected. As the display data, “reserved” in the display data string HDR is selected. In this case, the control signals that can be uttered are a VTR, a deck (audio cassette tape deck), an air conditioner (air conditioner) and the like in the control signal sequence SR.
The control signals indicated by “O” in the control signal sequence SR also include the corresponding electric device specifying information.
Here, there may be a case where a plurality of control signals can be generated corresponding to one input voice. For example,
When the operator pronounces "1", this indicates that the user has requested to set any one channel of the VTR, the television and the deck to "1". The specification of the target electrical device that can be transmitted with respect to the plurality of control signals is determined according to the transmission prohibition condition data (described in detail later) in the control signal-device state correspondence table 14 shown in FIG.

FIG. 5 is a diagram showing the device-word correspondence table shown in FIG. 4 in a general form. Words W1 to Wn, which are basic recognition results, are arranged in the kana string KR, and the display data string HD
In R, display data D1 to Dn are arranged corresponding to each word. Control signals to be generated are arranged under the corresponding electric devices M1 to MI corresponding to the words W1 to Wn. For example, a control signal von at which the word W1 is to be generated for the electric device M1 is arranged, a control signal ton is arranged at the intersection of the word W2 and the electric device M2, and the word Wn
A control signal aup is arranged at the intersection of the electric equipment M4. If there is no control signal to be generated for the word, a binary value "0" is placed. In FIG. 5, a control signal to be generated for electric equipment Mi corresponding to word Wn is represented by Mi [N].

FIG. 6 is a diagram showing an example of the contents stored in the device state memory 12 shown in FIG. The device status memory 12 includes a device name string AAR in which the names of electrical devices to be operated are arranged, and SDRs that store status data S [i] indicating the status of the corresponding electrical device corresponding to each electrical device. including.
If the state S [i] is 0, the power off state is set,
It is assumed that 1 indicates a power-on state, 2 indicates a specific operation 1 state, and 3 indicates another specific operation 2 state. According to the code of the state data S [i], the operation state of the corresponding device is set and monitored inside the remote control device.

FIG. 7 is a diagram showing a configuration of a working buffer memory used in the voice recognition unit 10 (second voice recognition unit 102) shown in FIG. This buffer memory is not explicitly shown in FIG.
Included in. In FIG. 7, the buffer memory stores the control signal Mi [x] at the address indicated by the pointer bufp. Here, x is a work variable. The buffer memory has J buffer registers, and the address of each buffer register is set by a pointer bufp.

FIG. 8 is a flowchart showing the operation of the voice recognition unit 10 shown in FIG. Hereinafter, the operation of the voice recognition unit 10 shown in FIG. 1 will be described with reference to FIG. First, the remote controller is initialized by inputting a voice through the microphone 2 or by a power-on switch (not shown). The voice recognition unit 10 sets the device status data S [i] stored in the device status memory 12 to all “0” states as initial settings. In this state, all the corresponding electric devices are in the power-off state (step S1).

In this initial state, the operator inputs information for operating the corresponding electric device by voice via the microphone 2. The speech recognition unit 101 recognizes the speech given via the microphone 2 and performs a recognition execution operation on the number of vocabulary words N stored in the device-word correspondence table 11 in advance. At this time, N words (corresponding to words W1 to WN) are also registered in the word standard pattern dictionary 113 shown in FIG. When the word recognized by the first speech recognition unit 101 is the second
Is given to the voice recognition unit 102. At this time, information obtained by coding the recognition result itself may be provided from the first speech recognition unit 101 to the second speech recognition unit 102,
Alternatively, only the word numbers in the word dictionary 113 (see FIG. 3) and the device-word correspondence table 11 may be provided to the second speech recognition unit 102. In FIG. 8, the first speech recognition unit 101 outputs word numbers (1- 1 in W1-WN) in the dictionary 113 or the device-word correspondence table.
N) x is given to the second speech recognition unit 102 (step S2).

When given the word number x from the first speech recognition unit 101, the second speech recognition unit 102 first sets the buffer register pointer bufp to 0, and sets the device data in the state data S [i]. The number i is set to 1 (step S3).

Next, the second speech recognition unit 102 accesses the device-word correspondence table 11 according to the word number x, and determines whether or not the corresponding control signal Mi [x] is 0 (step S4). ). Thus, a control signal to be generated corresponding to word number x is selected.

If a control signal Mi [x] other than "0" is found, this is a control signal to be generated from the remote control device, and the buffer register buf [bu shown in FIG.
fp + 1] is stored with the control signal Mi [x], and then the buffer register pointer bufp is incremented by one. As a result, first, the buffer pointer “1 (= bu
fp + 1) ”for the buffer register.
[X] is stored (step S5).

In step S4, if the control signal Mi [x] corresponding to the electric device i is 0, step S6
Proceed to.

In step S6, the electrical equipment name number i is incremented by one, and then it is determined in step S7 whether the value of i incremented by one exceeds the total number I of electrical equipment. If it is determined in step S7 that i is larger than I, it indicates that the control signals related to all the electric devices have been examined in correspondence with the recognized word (here, the electric device is Note that M1-MI).

If it is determined in step S7 that all the electric devices have not been searched yet, the process returns to step S4, and the control signal Mi located at the intersection of this electric device and the word number x (or word Wx). It is determined whether or not [x] is 0.

If it is determined in step S7 that i is larger than I, it indicates that all control signal searches for the recognition result Wx of the input voice have been performed, and the search result is stored in the buffer memory shown in FIG. Control signal Mi
Output all [x] to the device control unit 13 (step S
8). The control signal Mi sent to the device control unit 13
The number of [x] is given by the buffer pointer bufp.

Here, in the above-mentioned flow, the names and words of the electric devices are all represented by numbers, and the search operation is performed according to these numbers. At this time, in the device-word correspondence table, the words W1 to WI store the power-on words of the corresponding electric devices, and the WI + 1 to WI + 1 are stored.
A word for turning off the power of the related electric device is stored in the area of W2 · I. That is, words are grouped according to the operation state of each electric device, and words are stored in the device-word correspondence table.

Upon receiving all of the control signals from the voice recognition unit 10, the device control unit 13 changes the contents of the device state memory 12 and executes the device transmission prohibition condition data stored in the control signal-device state correspondence table 14. A transmittable control signal is searched, and the searched transmittable control signal is sent to the communication processing unit 3.

In step S2, display data is also retrieved from the device-word correspondence table 11 according to the word number x of the recognition result. The searched display data is transmitted to the communication processing unit 3. The processing of the display data will be described later.

Next, the details of the execution operation of the device control unit 13 in step S9 shown in FIG. 8 will be described.

FIG. 9 is a diagram showing an example of the contents stored in the control signal-device state correspondence table. In FIG. 9, the control signal-device state correspondence table 14 includes a control signal sequence SSR storing a control signal C [k] to be generated corresponding to the word number k.
Device name string A that stores the electrical device D [k] to be operated
A prohibition condition data string ISR storing the SR and the transmission prohibition condition E [k] of the device is included. Transmission prohibition condition data E
[K] is set in advance corresponding to the control signal C [k]. For example, if the control signal C [k] is "von" for turning on the power of the video tape recorder, it is not necessary to transmit the control signal von if the power of the video tape recorder is already on. At this time, in response to the control signal von, “1” indicating the video tape recorder as the target electric device is set as the electric device name D [k], and the transmission prohibition condition E of this video tape recorder is set.
“1” indicating the power-on state is set as [k].

In the transmission prohibition condition data E [k], for example, "prohibition of channel switching during video recording" is set. This is for example
The recognition word Wi is “1” indicating channel 1, the operation target electric device D [k] is set as a video tape recorder, and the corresponding transmission prohibition condition data E [k] indicates, for example, “2” indicating the recording operation state. It is realized by setting to. That is, the electric device D specified by the control signal C [k]
If [k] is in a certain operation state and the operation state matches the transmission prohibition condition E [k], the control signal C
Transmission of [k] is prohibited. Therefore, in the control signal-device state correspondence table 14, the electric device designated by the electric device D [k] stored in the electric device name column ASR is not necessarily the name of the electric device that actually receives the control signal C [k]. Note that they do not match. For example, a power off operation of the television during video recording is prohibited.

Next, the operation of the device control section 13 will be described with reference to FIG.

Here, the equipment control section 13 also
It is assumed that the buffer register shown in FIG. Each control signal is stored in the corresponding buffer register in the buffer register in the device control unit 13 by the register pointer in the same manner as the buffer register in the voice recognition unit 10.

The buffer register pointer is indicated by j. Pointer b used in the buffer register of FIG.
ufp indicates the number of storage control signals in the buffer register. The buffer register of the pointer j is indicated by buf [j]. 1 ≦ j ≦ J. The word number k in the control signal-device state correspondence table 14 satisfies 1 ≦ k ≦ K.

First, the device control unit 13 determines whether or not the pointer bufp of the buffer register is "0" (step S11). Buffer register pointer bu
When fp is “0”, the control signal to be transmitted is not stored in the buffer register, and the device control unit 13 ends the operation.

If it is determined in step S11 that the buffer register pointer bufp in FIG. 7 is not "0", the pointer j is set to 1 to select the first buffer register. As a result, the buffer register buf (1) is specified (step S12). Next, k is set to 1 in order to search for a prohibition condition in the control signal-device state correspondence table. Thereby, the control signal C [1] in the first row in the control signal-device state correspondence table,
The electric device D [1] and the transmission prohibition condition data E [1] are specified (step S13).

The control signal stored in the buffer register buf [j = 1] specified in step S12 is the control signal C [k = 1] included in the control signal-device state correspondence table.
Is determined (step S1).
4).

If a match is found in step 14, the device control section 13 searches the device status memory 12 for status data S [D [k]] of the electrical device D [k = 1]. Whether or not the device status data S [D [k = 1]] retrieved from the device status memory 12 matches the transmission prohibition condition data E [k = 1] stored in the control signal-device status correspondence table 14. Is determined (step S15).

In step S15, the device status data S [D [k = 1]] is replaced with the communication prohibition condition data E [k = 1].
If the control signal C [k =
1] is a control signal to be transmitted, and the control signal C [k = 1] is transmitted to the communication processing unit (step S16). The processing performed by the communication processing unit will be described later.

After outputting the control signal C [k = 1] to the communication processing unit, the device control unit 13 outputs the control signal C [k = 1].
The value of the status data S [D [k = 1]] stored in the device status memory 12 is changed in accordance with. That is, for example, when the control signal C [k = 1] is a power-on signal,
This state data S [D [k = 1]] is set to 1, and in the case of a power-off signal, the state data S [D [k =
1]] is set to “0”.

In step S14, if the control signal stored in the buffer register buf [j] does not match the control signal C [k = 1] stored in the control signal-device state correspondence table 14, this control signal-device In order to search for a corresponding control signal in the state correspondence table 14, the value of k is incremented by 1 (step S18).

The value of k is the control signal-device state correspondence table 1.
Then, it is determined whether or not the number of words is larger than the number K of words stored in the step S4 (step S19). A search for a corresponding control signal in state correspondence table 14 is performed.

If the value of k is larger than the value of K in step S19, the buffer register buf
This indicates that the control signal stored in [j = 1] was not searched in the control signal-device state correspondence table 14,
Returning to step S14, the operation of discriminating a match or a mismatch is performed.

If the value of k becomes larger than K in step S19, this is determined by the buffer register buf.
This indicates that the search operation to the control signal-device state correspondence table 14 for the control signal stored in [j] has been completed, and the process proceeds to step S20 where the buffer register buf
The value of the pointer j of [j] is incremented by one.

In step S15, the status data S [D [k]] stored in the device status memory 12 is set to the corresponding transmission prohibition condition data E [k] stored in the control signal-device status correspondence table 14. If they match, step S2
Move to 0. In this case, the buffer register buf [j]
Since the control signal stored in the buffer register is not transmitted, the process similarly proceeds to the next buffer register. Further, step S
Step 17 is also performed, and after the status data S [D [k]] in the device status memory 12 is changed in accordance with the searched control signal C [k], the process proceeds to step S20.

When the value of the next buffer register pointer j is incremented by one in step S20, it is determined whether or not the value of the buffer register pointer j exceeds the number bufp of the control signals sent from the speech recognition unit 10. A determination is made in step S21. The buffer register pointer bufp is a register pointer of a working buffer register used in the speech recognition unit 10, and the value of the register pointer bufp is the device control unit 1.
Given to 3. This may be configured such that the device control unit 13 simply searches the register pointer of the working buffer register stored in the speech recognition unit 10.

In step S21, when the value of the register pointer j is equal to or smaller than the number of transferred control signals bufp, it indicates that there is a control signal to be generated. Returning to step S13, the control signal-device state correspondence table 14 is again searched for the corresponding control signal C [k] and whether transmission is possible or not is determined.

In step S21, the value of the buffer signal pointer j is set to the value bu of the control signal to be transmitted.
If fp is larger than fp, it is determined that all control signal searches have been completed, and the operation of the device control unit 13 ends. Here, an example of a case where there are a plurality of control signals to be transmitted simultaneously is, for example, a channel setting signal, which is a video tape recorder (VTR), a television and an audio as shown in FIG. This can occur for tape decks.
In this case, the control signal is different for each electric device. Therefore, in such a case, there are a plurality of control signals to be transmitted.

In the above description of the operation, the device control section 13 is given a control signal from the speech recognition section 10 and stores the given control signal in a work buffer register provided therein. doing. However, in this configuration, the control signal storage buffer register may be shared by the voice recognition unit 10 and the device control unit 13. Next, the operation of the communication processing unit 3 will be described with reference to FIG.

The communication processing unit 3 is provided with a buffer register (not shown), and the transmission enable control signal from the device control unit 13 is stored in this buffer register. The buffer register pointer is given by m, and the value of the buffer register pointer m indicates the number of control signals given from the device control unit 13. First, it is determined whether or not the number of control signals provided from the device control unit 13, that is, the value of the buffer register pointer m is 0 (step S30). When the number m of control signals provided from the device control unit 13 is 0, the communication processing unit 3 does not perform any communication operation.

Next, if it is determined in step S30 that the value of the pointer m is not 0, the communication processing unit 3
Further determines whether or not the value of the pointer m is 1 (step S31). When the value of the pointer m is 1, there is only one control signal to be transmitted, and in this case, there is only one operation target electrical device, so the recognition result given from the voice recognition unit 10 (see FIG. (Display data shown in FIG. 4) is provided to the communication processing unit 3, and this display data is presented to the operator via the display unit 4 (step S3).
2).

The operator looks at the recognition result displayed on the display unit 4 and, when a control signal corresponding to the input voice information is indicated, an OK switch 5b included in the switch unit 5
A transmission permission signal is given by pressing (step S33). If the recognition result presented via the display unit 4 is incorrect, the control unit cancels the control signal by pressing the cancel switch 5a included in the switch unit 5 (step S33).

When the transmission permission is given in step S33, the communication processing section 3 transmits this control signal (step S34).

On the other hand, if the cancel switch 5a is depressed in step S33, the communication processing unit 3 invalidates the given control signal and does not transmit the control signal (step S35).

If it is determined in step S31 that the value of the pointer m indicating the number of control signals is greater than 1, the communication processing unit 3 presents the operable electric device name to the operator, and Is urged by the operator (step S4).
0). The display of the operable electric device name to the operator in step S40 may be performed via the display unit 4 shown in FIG. 1 or may be performed via the voice synthesizing unit 60 and the speaker 6. In this case, only one of the visual information and the audio information may be used, or both may be used. The configuration of the voice synthesizer 60 is arbitrary.

The presentation of the operable device name in step S40 is performed as follows. The control signals each include information for identifying an electric device. For example, in the control signal von for turning on the video tape recorder, the code of “v” indicates the video tape recorder. Therefore, the communication processing unit 13 presents the operable device name to the operator by viewing the electric device name code. Here, if there are multiple operable devices,
In addition to the above-described channel switching, there is "reservation". In the word "reservation" shown in FIG. 4, the reservation operation can be performed on a video tape recorder, a tape deck, and an air conditioner. In this case, since only one electric device is subjected to operation control, it is necessary to specify an electric device to be operated. Therefore, in this case, the display unit 4
Displays "Is electrical equipment A or electrical equipment B?". The same content may be pronounced via the voice synthesis unit 60 and the speaker 6.

The operator responds by inputting the name of the electric device to be operated by voice (step S41). The voice information input by the operator includes only the name of the electric device. The voice recognition unit 10 includes a word dictionary of the name of the electrical device, and provides the electrical processing name information corresponding to the input voice information to the communication processing unit 3 (this route is not shown). In this case, a word and display data are stored in the device-word correspondence table 11, and “0” is set so that no control signal exists in the corresponding control signal column SR.
Are arranged, and the voice recognition unit 10 may search for the identification electrical device data based on the recognition result of the input voice information.

The communication processing unit 3 specifies an electric device to be operated based on the electric device specifying information given from the voice recognition unit 10 (step S41). The specification of the operating device is performed, for example, by comparing the device code given from the voice recognition unit 10 with the device specifying code of each transmittable control signal, as described above. When the device to be operated is specified, the communication processing unit 3 presents the specified device in order to ask the operator for confirmation (step S42). The presentation of the specified device is displayed on the display unit 4. At this time, the presentation may be given to the operator by voice information via the voice synthesis unit 60 and the speaker 6.

The operator confirms the name of the electric device presented in step S41, and when the selected electric device is presented, presses the OK switch 5b of the switch section 5 to give a transmission confirmation signal. (Step S
43).

When the cancel button switch 5a of the switch section 5 is depressed in step S43, step S40
Then, the operator is presented with a plurality of electric device names again. If the OK button switch 5b is pressed in step S43, the process proceeds to step S34, where a control signal corresponding to the selected electric device is transmitted. In the above-described configuration, in the initial state, under the control of the voice recognition unit 10, the electric device state data S
All [i] are set to a state of “0” indicating a power-off state. However, there may be a situation in which, for example, after a power-on operation of a certain electric device is manually performed, the operation of the electric device is remotely controlled. In order to cope with such a situation, after the voice input of “initialization”, the corresponding operation state of the electric device is input by voice, and the recognition result of the input voice is written to the device status memory 12. The second voice recognition unit 102 may be configured as described above.

The form and communication method of the signal sent from the communication processing section 3 may be any as long as they can individually and independently control a plurality of electric devices.

The communication processing unit 3 may transmit a corresponding control signal without receiving confirmation from the operator as in steps S33 and S43 shown in FIG.

Further, in the flow shown in FIG.
In step S14, the control signal C is sent to the communication processing unit.
After [k] is output, the contents of the device state memory 12 are rewritten according to the control signal C [k]. At this time, step S17 is performed after the control signal C [k] is actually transmitted from the communication processing unit 3. Therefore, if the invalidation of the control signal is designated in step S35 in the operation flow shown in FIG. 11, the state data S [[Dk]] corresponding to the invalidated control signal is not changed.

Further, the voice recognition unit 10 and the device control unit 1
3. Each communication processing unit 3 is a microprocessor (MPU)
May be included, and each operation may be realized by one MPU.

[0081]

As described above, according to the present invention, the equipment
A word correspondence table and a control signal-device state correspondence table are provided, and a device state memory for storing an operation state of each electric device is provided, and a control signal corresponding to a recognition result of input voice information is provided to the device-word Since the control signal is transmitted only when the control signal retrieved from the correspondence table can be transmitted based on the device state memory and the control signal-device state correspondence table, the electric power to be operated is It is not necessary to switch between the speech recognition result and the conversion map for the device control signal for each device, and the conversion map IC is used for each electrical device.
There is no need to save the card or provide a switch to switch the conversion code for each electrical device to be operated,
A voice recognition remote controller with excellent operability can be obtained.

Further, if there are a plurality of control signals that can be transmitted and it is not possible to specify the electric device to be operated, the name of the operable electric device at the time of operation is presented, and the name of the electric device is displayed to the operator. Since it is configured to promote the specification, it is possible to obtain a voice recognition remote control device which is excellent in a backup system in a case where the operation target cannot be identified, and which can surely remotely control a desired electric device.

[Brief description of the drawings]

FIG. 1 is a block diagram functionally showing the overall configuration of a voice recognition remote controller according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of an external appearance of a voice recognition remote controller according to an embodiment of the present invention.

FIG. 3 is a diagram illustrating an example of a specific configuration of a first speech recognition unit included in the speech recognition unit illustrated in FIG. 1;

FIG. 4 is a diagram showing an example of a storage form of a device-word correspondence table shown in FIG.

FIG. 5 is a diagram showing the contents stored in the device-word correspondence table shown in FIG. 4 in a coded form.

FIG. 6 is a diagram showing an example of contents stored in a device state memory shown in FIG. 1;

FIG. 7 is a diagram illustrating an example of a configuration of a working buffer register used in the voice recognition unit.

FIG. 8 is a flowchart showing an operation of the voice recognition unit shown in FIG. 1;

9 is a diagram showing an example of the contents stored in a control signal-device state correspondence table shown in FIG. 1;

FIG. 10 is a flowchart showing the operation of the device control unit shown in FIG. 1;

FIG. 11 is a flowchart showing the operation of the communication processing unit shown in FIG. 1;

[Explanation of symbols]

 2 Microphone 3 Communication processing unit 4 Display unit 5 Switch unit 6 Speaker 10 Voice recognition unit 11 Device-word correspondence table 12 Device status memory 13 Device control unit 14 Control signal-device status correspondence table 60 Voice synthesis unit 101 First voice Recognition unit 102 Second speech recognition unit

Claims (2)

(57) [Claims] 1. A voice recognition remote control device for remotely controlling a plurality of electric devices by an input voice, wherein the input voice includes control target electrical device information and control operation information, and recognizes the input voice. First to output recognition result
Device / word correspondence storage means for storing a plurality of basic speech recognition results and a control signal for controlling an electric device to be generated corresponding to each of the basic speech recognition results in association with each other, and the electric device Device state storage means for storing information indicating the operating state of the control device, control signal / device state correspondence storage means for storing the control signals and the control target electric devices in association with each other, and the control signal / device state correspondence storage means Stores a transmission prohibition condition flag, which is generated in advance according to the operation state of each electric device and indicates whether a control signal can be transmitted, in association with each of the stored control target electric devices, and stores the first voice recognition means. Second speech recognition means for searching a corresponding control signal from the device / word storage means based on the recognition result from the control signal / device based on the searched control signal A transmission corresponding to the control target electric device and the associated transmission prohibition condition flag are searched from the state correspondence storage means, and when the retrieved transmission prohibition condition flag indicates that transmission is possible, the transmission for transmitting the searched control signal is performed. A voice recognition remote control device comprising means. 2. The voice recognition remote control device according to claim 1, wherein when a transmission prohibition condition flag associated with the searched control signal indicates that transmission to a plurality of controlled electrical devices is possible, the operator is notified. Means for presenting the names of the plurality of control target electrical devices and stopping transmission of the control signal are further provided.