JPH01193722A - Speech recognition camera - Google Patents

Abstract

PURPOSE:To confirm the operation state of a speech recognition means by displaying the completion of the registration or recognition of an instruction voice. CONSTITUTION:When an E switch 16a is turned on, an ES display part 18a blinks and 'exposure' is inputted at this time. A speech recognizing circuit 3 registers the 'exposure' as an instruction voice for setting an exposure correction mode. Thus, the instruction is registered and then the ES display part 18a stops blinking and illuminates, and when the E switch 16 is turned off, the display part turns off. Similarly, instruction voices are registered corresponding to respective switches which constitute a registration switch 16. At this time, an IT display 18c, a numeric display 18d, and overexposure and underexposure displays 18h and 18k blink, illuminate, and turn off similarly to the ES display 18a and ME display 18b. Consequently, a user can confirm that the respective instruction voices are registered.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a voice recognition camera operated by voice input.

(Prior art) With the advancement of voice recognition technology, voice recognition cameras are being developed in which when a pre-registered voice or keyword (command voice) is input, the camera recognizes the voice and performs a predetermined action. ing. If the camera is operated by voice, the user can release the shutter by simply saying, ``release,'' for example, without worrying about single camera shake, and can operate the camera remotely.

By the way, in the voice recognition cameras that have been proposed so far, it is not possible to know whether the voice has been registered or recognized, so if there is a registration error or if the voice does not work even if the voice is input, the voice recognition I don't know if there is something wrong with the system or if something is wrong with the camera, and
At present, there is no mention of this type of problem.

(Purpose and Structure of the Invention) The present invention has been made in view of the above points, and has the purpose of making it possible to check the operating state of a voice recognition system, and for that purpose, registering or recognizing a command voice. A display means is provided to display this information.

(Embodiment) FIG. 1 is a block diagram showing a circuit configuration of an embodiment of a voice recognition camera according to the present invention. 3 is an amplifier that amplifies the input voice signal; 3 is a voice recognition circuit that determines whether the input voice signal matches the registered real voice, and when it matches, outputs a signal corresponding to the voice; 4 is a microcomputer that controls each part of the camera according to the output signal from the voice recognition circuit 3;
Reference numeral 5 denotes a photometry circuit that measures the subject brightness using a photodiode 6; 7 a shutter control circuit that controls the shutter speed by turning on and off the magnet 8 according to the subject brightness; 9 detects the focus of the photographic lens and moves the photographic lens to the focal position. AF circuit to move the film to
This is a motor drive circuit that drives the.

Next, to explain the camera operation switches, 12.13
S, , S, which are turned on by the release button, respectively.
When the release button is pressed down one step lightly, the S switch 12 is turned on, and when the release button is pressed down further, both the S switch 12 and the 82 switch 13 are turned on. In the camera shooting sequence, when S and switch 12 are turned on, light metering and distance measurement are performed in order, and S2 switch 1 is turned on.
When 3 is turned on, lens drive, shutter drive, and film winding are performed. 14 is a self-timer switch that operates a self-timer, and 15 is a rewind switch that is operated when rewinding the film.

16 is a registration switch for registering the command voice;
This switch is constituted by a matrix switch and is connected to the voice recognition circuit 3.

The registration switch 16 includes a switch for registering a command voice for setting a shooting mode, and a switch for registering a command voice for setting shooting conditions during the mode.

@Person uses the E switch 16 to register the exposure compensation command voice.
a. M switch 16b for registering multiple exposure command voice;
and an I switch 16c that registers a command voice for interval exposure, the latter of which is a + switch 16d that registers a command voice to set overexposure when in exposure compensation mode.
, a switch 16e for registering a command voice to set the output under, and a numeric switch 16f for registering a command voice to set the aperture correction amount, the number of films for multiple exposure, or the interval time for interval exposure in each mode. . As a registration switch, there is also an F switch 16 that registers the focus lock command voice.
g. S switch 16 for registering shutter operation command voice
h. ST switch 1 for registering command voice for self-timer
6h, and an R switch 16M for registering a voice command for film winding (rewinding) operation. When the registration switch 16 is operated, the voice recognition circuit 13 enters the voice registration mode, and at this time, all camera operations other than the display section 18 are prohibited. Furthermore, when the registration switch 16 is not operated, the voice recognition circuit 13 is in voice recognition mode. These registration switches register the command voices shown in the table below, for example.

(The following is a blank space) 17 is a display unit driving circuit that drives the display unit 18 according to the output from the voice recognition circuit 3 or the microcomputer 4.

The display section 18 is, for example, numbered at an easily visible position on the top surface of the camera, and is configured as a liquid crystal display or the like as shown in FIG. 2.

In FIG. 2, reference numeral 18a denotes an ES display that lights up when an exposure correction command voice is registered by the voice recognition circuit 3 during registration mode and when recognized during recognition mode.
8b is an IT display section 18c which lights up when a command voice for multiple exposure is registered and recognized; 18c is an IT display section which lights up when a command voice for interval exposure is registered and recognized;
8d is a numeric display that displays the number when the numeric command voice is registered and recognized, and 18e, 18f, and 18g are time displays that display the time when the interval time command voice is recognized. It represents hours, minutes, and seconds in that order. 18h is an overexposure display that lights up when a command voice for overexposure setting is registered and recognized in exposure compensation mode, and 18h is an underexposure display that lights up when a command voice for underexposure setting is registered and recognized. It is.

Also, 18M is an S1 display that lights up when a command voice for focus lock is registered and recognized, and 18m is an S1 display that lights up when a command voice for shutter operation is registered and recognized.
2 display, and at this time, not only S2 display 18m but also S
8 display 1841 also lights up. 18n is a dialogue display that lights up when the command voice for self-timer operation is registered and recognized, and 18p is a rewind display that lights up when the command voice for rewind operation is registered and recognized.

The numerical display 18d not only displays the shooting conditions for special photography, but also displays the number of shots of the film.

Next, registration of command voices will be explained using the case of special photography as an example.

First, when you turn on the E switch 16a, the ES display section 18a
flashes. At this time, when inputting ``Rosetsu'', the voice recognition circuit 3 registers ``Rosetsu'' as a command voice for setting the exposure correction mode. Once registered, the ES display section 18a stops blinking and lights up, and then the E switch 16a is turned on.
When the switch is turned off, the light goes out. When the M switch 16b is turned on, the ME display section 18b is dimmed. At this time, when the voice ``Taju'' is input, it is registered as a command voice to set the multiple exposure mode. . When registered, the ME display section 18b stops blinking and lights up, and then turns off when the M switch 16b is turned off. Similarly, command voices (see the above table) are registered in correspondence with each switch constituting the registration switch 16. At this time, IT display 18c, numeric display 18d, "il output over and under display 1
8h, 18 also have the ES display 18a and ME display 18 mentioned above.
Blinks, lights up, and turns off in exactly the same way as b, and also has S1 display 181, S2 display 18m, and self display 18n.
, the rewind display tap operates similarly. These displays allow the user to confirm that each command voice has been registered.

Next, the operation in the recognition mode will be explained. When performing special photography, a command voice is input from the microphone 1 to set the photography mode and its conditions.

Now, suppose you want to stop down by 2 stops and take a picture with slightly underexposure, input a sound into the microphone L, and first set the shooting mode. When "Roschutsu" is recognized, the ES display 18a lights up, allowing you to confirm that it has been recognized.

After recognizing the mode, the voice recognition circuit 3 becomes ready to recognize the shooting conditions, and then when "r under" is input, the underexposure display 18k lights up to indicate that "under" has been recognized. Enter "2" into the microphone L.

When the voice recognition circuit 3 recognizes "2", the number display 18d displays "2". The setting of the shooting mode and shooting conditions is now complete, and the voice recognition circuit 3 outputs a signal according to the set mode and conditions to the microcomputer 4, and the microcomputer 4 controls each part of the camera accordingly.
When shooting begins, the image will be slightly underexposed.

Similarly, when inputting "Tajuu" into the microphone L, the ME display 18b lights up, and then "
If you enter "Zero" or "Hachi", "08" will appear on the number display 18d.
is displayed, and the voice recognition circuit 3 recognizes that this is eight multiple exposure shots.

Regarding the interval exposure mode, when "interval" is input, the IT display 18c lights up and it is recognized that the mode is the interval exposure mode. After that, the voice recognition circuit 3 enters a state where it accepts the interval time, and continues to say "zero", "ichi", "ichi", "go", "zero".
, when inputting "zero", roll is displayed on the hour display 18e, r15J is displayed on the minute display 18f, and "00" is displayed on the second display 18g, indicating that the interval time of 1 hour and 15 minutes has been recognized. Furthermore, to set the number of shots, select "Zero", r
When inputting ``San'', ``03'' is displayed on the number display 18d. The recognition results of the voice recognition circuit 3 are output to the microcomputer 4, and the microcomputer 4 controls each part of the camera accordingly.
The camera repeats three shooting operations every 1 hour and 15 minutes.

The same operation applies to other displays. In other words, “
When "focus lock" is recognized, the S1 display 1841 lights up, and when "release" is recognized, both the S2 display 18 sentence and the SO display 18m light up. In addition, “self” and “
When “Rewind” is recognized, each self-display 18
n, the rewind display 18p lights up.

Each of the above displays goes out when each camera operation ends.

In the above embodiment, a registration mode and a recognition mode are provided as the operation modes of the voice recognition circuit 3, but a confirmation mode may also be provided in addition.If the camera is not used for a long period of time, the command voice You may want to check whether the camera can recognize you or whether the input voice is registered. In such a case, set the voice recognition circuit to confirmation mode, and in this mode, the camera will not operate even if the command voice is input, but the recognition result will be displayed on the display section 18. good.

By the way, FIG. 3(A) shows the installation location of the microphone used in the voice recognition camera, and the first microphone 20 for voice input is installed at the lower part of the back cover 21 of the camera. This microphone may be installed at the bottom of the camera.In short, when the photographer issues a command voice while looking through the viewfinder 22, the microphone 20 is placed in a position where the voice can be easily captured.
is installed.

Furthermore, since the command voice may be emitted from in front of the camera (for example, when taking a group photo), there is a third
As shown in Figure (B), a second microphone 23 is installed. The first microphone 20 and the second microphone 23 can be switched by a changeover switch 24. However, when the camera is in the registration mode, that is, when audio registration is being performed by the registration switch 25, registration is accepted from the first microphone 20, and the second microphone 23 is turned off.

As the second microphone 23, one having directivity toward the front is used, and the directivity corresponds to the shooting angle of view. In other words, voice input is accepted only when a voice command is given within the shooting angle of view.

On the other hand, in a camera having a zoom function, the directivity of the microphone can be switched depending on the edge of the lens. For example, as shown in FIG. 4(a), a microphone 27 movable in the optical axis direction of the photographic lens is provided in the microphone hole 6a of the camera body 26, and the microphone 27 and the zoom lens 28 are connected by a lever 29. Link. As shown in FIG. 4 (b), when the zoom lens 28 is extended (for tele), the microphone hole 26
Move the microphone 27 toward the back of a. As a result, the directivity of the microphone 27 is narrowed, and the directivity of the microphone 27 can be changed in response to a change in the photographing angle of view.

(Effects of the Invention) As explained above, in the present invention, the fact that the command voice has been registered or recognized is displayed, so it is possible to confirm the operating state of the voice recognition means. In addition to being able to prevent this, even when voice input does not work, it is possible to know whether the voice recognition system is at fault or the camera is at fault.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the circuit configuration of one embodiment of the voice recognition camera according to the present invention, Fig. 2 is an enlarged view of the display section of the camera, and Figs. FIGS. 4A and 4B are installation diagrams of an audio input microphone in a zoom lens camera.

Claims (1)

[Claims] A voice recognition camera that recognizes a command voice and operates in accordance with the voice recognition camera, characterized in that the voice recognition camera is provided with a display means for displaying that the command voice has been registered or recognized.