JP2001350492A - Speech recognition faucet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a speech recognition faucet which is constituted to activate its main body function section by the user's speech and is well activated by surely recognizing the user's speech. SOLUTION: This speech recognition faucet is constituted to collate the speech data of the instruction by the speech emitted by the user with the speech data previously stored in the memory section of a control section and to control the activation of the main body function section 10 of the faucet by the control section. The speech data of unspecified plural persons is previously stored as standard data for collation by each of the instruction contents into the memory section of the control section. The speech data by the faucet user and the standard data are collated in the control section and if there is the corresponding speech data, the activation of the corresponding instruction contents is effected by the main body function section 10. On the other hand, if there is no corresponding speech data, the user's speech data is additionally registered in the memory section and is used as the subsequent speech data for collation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition faucet which recognizes a user's voice and performs an operation of an instruction content by the voice.

[0002]

BACKGROUND OF THE INVENTION Voice data of an instruction issued by a user is compared with voice data stored in a memory unit of a control unit, and the control unit controls the operation of a faucet function unit, for example, a valve unit. 2. Description of the Related Art Voice recognition faucets for discharging water, supplying hot water, or stopping water are conventionally known.

[0003] The following voice recognition techniques are used for this type of voice recognition faucet: (A): control is performed by taking voice data of a large number of unspecified people of several hundreds for each instruction content. Is stored in advance in the memory unit of the unit, and the data is stored as standard data for comparison (for data having a common feature (the greatest common denominator feature) among a large number of audio data), these are collectively collected into a single or The same shall apply hereinafter), and the voice data generated by the user when the faucet is used is checked against each of these standard data to control the instruction content of the corresponding voice data. One that is executed by the functional unit under control of the unit may be considered.

In the case of the voice recognition faucet of (A), since a large amount of voice data of an unspecified person is stored in the memory unit as standard data for comparison, the voice recognition rate of the user is high. However, there is a problem that the required cost is inevitably increased because a large amount of audio data must be obtained and processed.

[0005] (B): For example, it is considered that voice data of typically an adult man, an adult woman, and a child are stored in advance in a memory unit of the control unit, and this is used as the standard data for collation. Can be

The voice recognition faucet (B) has the advantage that the cost is low because the amount of voice data to be stored in the memory unit is small, but the standard data amount for collation is small. Inevitably, there is a problem that the recognition rate of the user's voice is low. That is, there is a problem that the accuracy of voice recognition is poor.

On the other hand, (C): the faucet is used when installing the faucet without storing in advance the voice data of an unspecified person in the memory of the control unit as standard data for verification. It is also conceivable that the user's own voice data is stored (registered) in the memory unit of the control unit, and this is used as voice data for comparison.

In the case of the voice recognition faucet of the form (C),
There is no need to collect the voice data of an unspecified person in advance and create standard data for verification, thus reducing the cost required for that purpose. However, it is necessary to temporarily register the user's own voice data in the memory unit of the control unit, which is troublesome for the faucet user.
If only the voice of a human user is registered, only that person can operate the faucet by voice, and there is a problem that many other users cannot activate the faucet by voice.

[0009]

The voice recognition faucet of the present invention has been devised to solve such a problem. According to the first aspect of the present invention, the voice data of the instruction issued by the user is collated with the voice data stored in the memory unit of the control unit, and the control unit controls the operation of the functional unit of the faucet. In the voice recognition faucet, the voice data of a plurality of unspecified persons is stored in advance in the memory unit of the control unit as standard data for collation for each instruction content, and the memory unit is further stored in the memory unit. An additional registration area for new voice data of the user is provided in the control section, and the voice data of the user of the faucet is compared with the standard data by the control unit, and if there is the corresponding voice data, the corresponding instruction content is provided. While the operation is performed by the function unit, when there is no corresponding voice data, the voice data of the user is additionally registered in the additional registration area and used as voice data for subsequent verification. .

According to a second aspect of the present invention, in the first aspect, the control section controls the operation of a valve section in the function section.

According to a third aspect of the present invention, in any one of the first and second aspects, the control unit executes a voice registration mode for additionally registering the voice data of a user based on a switch-on operation. It is characterized by being done.

[0012]

As described above, in the voice recognition faucet of the present invention, the voice data of a plurality of persons is stored in advance in the memory unit of the control unit for each instruction as standard data for collation, and If the user's voice data corresponds to any of the standard data, the faucet is activated based on the standard data, while if the standard data does not correspond to the user's voice data, the user's voice data is used. The voice data is additionally registered (stored) in the memory unit again, and is used as voice data for subsequent verification. In other words, the speech recognition faucet of the present invention is a faucet of the above (A) or a combination of (B) and (C). There is no problem that the faucet does not recognize and the faucet does not work under voice.

If the faucet user's voice data corresponds to pre-stored standard data for verification,
When installing the faucet, there is no need to perform the process of registering the voice data of the user in the memory unit. There is no trouble that the tap user has to do it himself.

In the voice recognition faucet of the present invention, when the user's voice data does not correspond to the standard data,
Since the voice data of the user can be additionally registered and used as voice data for verification, it is not necessary to store a particularly large amount of data as standard data for verification. That is, it is only necessary to store, for example, voice data of typically adult men, adult women, children, and the like as standard data, thereby reducing the required cost for collecting and processing the standard data. That is, in the present invention, the voice recognition faucet can be suitably configured in a form in which the faucet of the above-described type (B) and the faucet of the type (C) are combined.

In the present invention, for example, when an instruction for water is given by voice, when a foreigner is a user, additional registration can be made by saying "water" instead of "water". There is an advantage that when the same user subsequently says "water", this can be recognized and the water can be discharged.

In the present invention, the operation of the valve section in the functional section can be controlled by the control section. In this case, for example, the water side valve is opened by the user's "water" or similar meaning voice data to discharge water, and the hot water side valve is opened by "oyu" or similar meaning voice data. The hot water can be discharged, and the hot water side valve or the water side valve can be closed with audio data meaning "stop" or similar to stop the hot water or water discharge.

In the present invention, the control unit can execute a voice registration mode in which voice data of the user is additionally registered based on a switch-on operation by the user (claim 3).

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a voice recognition faucet (automatic faucet) of the present embodiment. In the figure, reference numeral 10 denotes a main body function unit disposed below a counter or the like, and 12 denotes a main body functional unit disposed above the counter or the like. It is a spout section to be installed. Here, the main body function unit 10 is attached and fixed to a wall via a leg tube 14.
Reference numeral 16 denotes a flow control plug provided at the base of the leg tube 14.

The main body function unit 10 comprises a mixing unit 18 for mixing water and hot water supplied through a pair of leg pipes 14 at a predetermined ratio, and water or hot water or mixed water from the mixing unit 18 (hereinafter simply hot water). A water-side valve portion 20 and a hot-water-side valve portion 21 (see FIG. 2) for supplying and shutting off water to and from the water discharge portion 12.
A control unit 22 for controlling the operation of the valve units 20 and 21 is provided.

A temperature control handle (temperature control handle) 28 is connected to the main body function unit 10 via an operation cable 26.
Is connected, and by operating the temperature control handle 28, the adjusted temperature in the mixing unit 18 is set or changed. A power cord 30 also extends from the main body function unit 10, and power is supplied to the main body function unit 10 by inserting a power plug 32 at the end thereof into an outlet.

The water discharge section 12 has a water discharge head 36 having a water discharge port 34 at a tip end thereof.
6 is provided with a manual water discharge / stop switch 38. The manual water-discharge switch 38 also serves as an operation switch for registering the voice of the faucet user, as will be described in detail later. A human body detection sensor 42 and a microphone 44 are provided on the upper surface of the water discharge pipe 40. Here, a pyroelectric sensor is used as the human body detection sensor 42.

The manual water-discharge switch 38, the human body detection sensor 42, and the microphone 44 provided on the water discharge unit 12 are connected to the control unit 22 of the main body function unit 10 by a conducting wire 46. Further, the water discharge pipe 40 and the main body function unit 10, more specifically, the outlet of water or hot water are connected by a conduit 48, and water or hot water from the main body function unit 10 is sent to the water discharge pipe 40 through the conduit 48. I have.

As shown in FIG. 2, the control section 22 has a central processing section 50 and a memory section 52, and the memory section 52 includes, as typical voice data, adult men, adult women, and children. Each of the two sets of voice data is stored in advance as standard data for collation for each instruction content.

In this voice recognition faucet, when the user of the faucet gives an instruction by voice, for example, when the user speaks “water”, the voice signal is transmitted to the central processing unit 50 via the microphone 44 and voice recognition circuit 54. Is input to the central processing unit 50
Is compared with the standard data stored in the memory unit 52 in advance, and when there is audio data corresponding to the standard data, the operation corresponding to this is performed by the main body function unit 10, specifically, the valve units 20, 21.

For example, when the faucet user speaks "water" to produce water, that is, when the user gives an instruction to produce water by voice, the voice data of the user is processed by the central processing unit 5.
0 is sequentially collated with audio data stored in advance in the memory unit 52 as standard data, and when it is determined that any of a plurality of audio data stored in the memory unit 52 is satisfied, the water side valve unit 20 Is opened to discharge water from the water discharge port 34.

More specifically, when the faucet user is an adult male and gives an instruction by voice "mizu", the voice data is usually N in standard data for verification (standard data area X).
o. 1 (see FIG. 3), and the control unit 22 opens the water-side valve unit 20 via the water-side valve drive circuit 56 so that water is discharged from the water discharge port 34.

On the other hand, when an adult male utters "Oyu", the voice data is usually the standard data No. 2, and the control unit 22 responds accordingly.
The hot water side valve drive circuit 58 opens the hot water side valve portion 21 via the hot water side valve drive circuit 58 to discharge hot water from the water discharge port 34.

Further, when an adult male utters "stop", the voice data is usually the standard data No. Three
In this case, both the water side valve portion 20 and the hot water side valve portion 21 are closed, and the outflow of water or hot water from the water discharge port 34 is stopped.

Similarly, when an adult woman utters "Mizu", or "Oyu", or "Stops"
When the voice is pronounced, the voice data is normally N data of the standard data.
o. No. 4 or No. 4 No. 5 or No. 5 6 and the control unit 22 determines that the newly input voice data and the standard data in the memory unit 52 match, ie, the corresponding operation, that is, the water side valve unit 20 The valve opening, the opening of the hot water side valve section 21 or the closing thereof are performed. Similarly, even when the child issues an instruction with the above content by voice, the same process is performed to cause the water-side valve portion 20 and the hot-side valve portion 21 to perform the operation of the corresponding instruction content. .

By the way, the manner or content of the faucet user's utterance is different for each individual. Therefore, even if the faucet user utters "Mizu", "Oyu" and "Stop",
The control unit 22 may not recognize this well.

In particular, in this example, typically only audio data of about two persons each of an adult man, an adult woman and a child is taken and stored as standard data. It can sufficiently occur that there is no corresponding audio data.

Therefore, in this example, if the faucet does not work as instructed by the faucet user's voice after the faucet is installed, a new memory unit 52 (additional data area (additional registration area) Y in FIG. ) Additionally registers (stores) the voice data of the faucet user, and thereafter uses this as voice data for verification.

For example, when the faucet user speaks “water” in his / her own voice, the voice data is registered in the additional data area Y in the memory unit 52 (No. K), and when the faucet is pronounced “yu”. The voice data is registered in the additional data area Y (No. K + 1). Further, when "stop" is issued, the sound data is registered in the additional data area Y (No. K + 2). Thereafter, when the faucet user gives an instruction by voice, the voice can be surely recognized by the control unit 22, so that the valve units 20 and 21 of the main body function unit 10 can be reliably operated. Will be able to

FIG. 4 shows the details of the control by the control unit 22. As shown in the drawing, in the water faucet of this example, when the human body detection sensor 42 does not detect a human body, both the water side valve portion 20 and the hot water side valve portion 21 are kept in a closed state (step S1).
0, S12) On the other hand, when a human body is detected and the manual water discharge switch 38 is turned on, the manual water discharge mode is executed. That is, the water is discharged from the water discharge port 34 by the switch operation regardless of the presence or absence of the voice. When the same water discharge switch 38 is turned off, that is, when the water stop operation is performed, the water is stopped in the manual water discharge mode (steps S10, S14, S16).

On the other hand, when there is a voice input from the user under the human body detection state, the voice is stored (step S).
18, S20), and the input voice data is collated with the voice data stored in the memory unit 52.

More specifically, the input voice data is collated one by one with each voice data stored in the memory unit 52, and any one of the voice data in the memory unit 52 When there is, an operation corresponding to the main body function unit 10 is performed. More specifically, when the audio data has any of the following contents: “water”, “yu”, and “stop”, the water-side valve 20 is opened, the hot-water valve 21 is opened, or any of them. The valve is closed (steps S22 and S22).
24, S26, S28).

The above steps are repeated as long as the human body detection sensor 42 continues to detect a human body. When the user leaves the front of the water discharge section 12 and the human body detection sensor 42 stops detecting the human body, the water-side valve section is stopped. Both the hot water side valve unit 20 and the hot water side valve unit 21 are closed (steps S30 and S30).
32).

As described above, in this embodiment, the memory unit 52
When there is no data corresponding to the user's voice data in the standard data stored in advance in the memory unit 52, the user's own voice data can be registered in the additional data area Y of the memory unit 52.

The processing for this registration is performed in step S26.
When it is determined that there is no corresponding voice data in the memory unit 52, the voice registration mode of step S34 can be automatically executed.

However, after step S26, the voice registration mode (step S34) is not automatically executed, and in some cases, the user's intention is set separately from the execution of each step shown in FIG. , The voice registration mode can be executed independently of the steps shown in FIG.

After the execution of the voice registration mode, the voice of the user himself is stored in the memory unit 52. Therefore, after that, the voice data of the user is stored in the voice data of the memory unit 52. Since there is data corresponding to the data, there is substantially no transition from step S26 to step S34.

FIG. 5 shows an example of a procedure for registering a user's own voice. Here, when the water stop switch 38 is kept pressed for, for example, 6 seconds, a voice registration mode is entered.
The voice is synthesized (see FIG. 2), and an announcement “register the voice” is issued from the speaker 60 (step S36).
Next, following this, an announcement "Please say yourself" is also made from the speaker 60 (step S3).
8) In response, the user utters "Water" (step S40), and the voice of "Water" of the user is subjected to data processing and registered in the additional data area Y of the memory unit 52 (step S42). ).

After that, the speaker 60 is again notified that "water has been registered" (step S44), and the registration processing of the voice of "oyu" and the registration processing of the voice of "stop" are performed in steps S46 to S46. This is performed through each step of step S60. Finally, in step S62, an announcement that "sound registration ends" is announced, and a series of registration procedures ends.

As described above, with the voice recognition faucet of this embodiment, there is no problem that the faucet does not recognize the user's voice and the faucet does not work under the voice.

If the faucet user's voice data corresponds to pre-stored standard data for verification,
When installing the faucet, there is no need to perform processing for registering the voice data of the user in the memory unit 52. Therefore, in order to be able to use the faucet, the processing for once registering the voice of the faucet user must be performed. There is no trouble that the tap user has to do it himself.

In the voice recognition faucet of this embodiment, when there is no user data corresponding to the standard data, the voice data of the user can be registered and used as voice data for verification. Therefore, it is not necessary to store a particularly large amount of data as standard data for collation. Therefore, in this example, as described above, it is only necessary to store representative voice data of, for example, an adult man, an adult woman, and a child as standard data, thereby reducing the cost required for collecting and processing the standard data. It can be kept low.

Although the embodiment of the present invention has been described in detail, this is merely an example, and the present invention can be configured in variously modified forms without departing from the gist thereof.

[Brief description of the drawings]

FIG. 1 is a diagram showing an overall configuration of a voice recognition faucet according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a control system by a control unit according to the same embodiment.

FIG. 3 is a diagram illustrating storage contents of audio data in a memory unit.

FIG. 4 is a diagram illustrating a control content of a control unit of FIG. 2 in a flowchart.

FIG. 5 is an explanatory diagram showing an example of a voice registration procedure by a control unit in FIG. 2;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Main-body function part 20 Water side valve part 21 Hot water side valve part 22 Control part 38 Discharge water switch 52 Memory part X Standard data area Y Additional data area (additional registration area)

Claims (3)

[Claims] The present invention is characterized in that voice data of an instruction given by a user's voice is compared with voice data stored in a memory unit of a control unit, and the control unit controls the operation of the faucet function unit. In the voice recognition faucet, in the memory unit of the control unit, voice data of a plurality of unspecified persons is stored in advance as standard data for collation for each instruction content, and the memory unit further stores the user's voice data. An additional registration area for new voice data is provided, and voice data by the user of the faucet is compared with the standard data by the control unit, and if there is corresponding voice data, the operation of the corresponding instruction content is performed by the function unit. On the other hand, when there is no corresponding voice data, voice data of the user is additionally registered in the additional registration area and used as voice data for subsequent verification. 2. The voice recognition faucet according to claim 1, wherein the control section controls the operation of a valve section in the function section. 3. The control device according to claim 1, wherein the control unit executes a voice registration mode for additionally registering the voice data of a user based on a switch-on operation. Features a voice recognition faucet.