JPS6365497A - Voice generator - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a voice generating device such as an automatic announcement device.

[Prior Art] Conventionally, a sound generating device as described in Japanese Utility Model Application Publication No. 57-188500 has been proposed. This voice generating device includes a memory circuit that stores voice information as a digital signal, a voice synthesis circuit that reads voice information written in the memory circuit and reproduces the voice information as an analog signal, and an activation of the voice synthesis circuit. It is configured with a sensor that detects timing and a speaker that is activated by the output of a voice synthesis circuit.

[Problems to be Solved by the Invention] However, in the conventional voice generating device described above, each time the sensor senses an object of a person to whom voice is generated (such as a customer visiting a sales floor), the voice synthesis circuit is activated. A playback start command is output. Therefore, if the object to be sensed frequently moves back and forth in the sensing range of the sensor, the speaker will output audio information frequently, which may cause discomfort to the person to whom the audio is being generated, or may reduce the effectiveness. On the other hand, there are disadvantages such as halving.

In addition, in the above-mentioned sound generation device, without using a sensor,
It is also conceivable to operate the speech synthesis circuit repeatedly at preset playback intervals and output the speech information every predetermined period of time. However, in this case, the timing at which the voice information is generated is monotonously repeated, making it less appealing to the voice generation target.

An object of the present invention is to supply voice information at appropriate timing to a person to whom voice is to be generated.

[Means for Solving the Problems] The present invention provides a storage circuit that stores audio information as a digital signal, and an audio synthesis system that reads audio information written in the storage circuit and reproduces the audio information as an analog signal. In a sound generation device comprising a circuit, a sensor for detecting the activation timing of a speech synthesis circuit, and a means for outputting speech information of the speech synthesis circuit, a predetermined reproduction interval time is set, and the sensor detects the activation timing of the speech synthesis circuit. The apparatus includes a means for ignoring activation timing detected within the specified time and not using it for activation of the speech synthesis circuit.

[Operation] According to the present invention, the voice synthesis circuit is activated at the activation timing detected by the sensor, and after the voice information recorded in the storage circuit is outputted from the voice information output means via the voice synthesis circuit, All activation timings newly detected by the sensor within the regeneration interval time are ignored. Namely.

The next activation of the speech synthesis circuit is performed at the activation timing newly detected by the sensor after the reproduction interval time has elapsed. Therefore, even if the object to be sensed, such as the person to whom the sound is generated (customer visiting the sales floor, etc.), activates the sensor frequently, the sound information will not be repeatedly output, and the next sound information Since the output is based on the detection result of the sensor, the voice information is supplied to the person to whom the voice is generated at an unexpected and appropriate timing.

[Embodiment] FIG. 1 is a block diagram showing an example of a voice recording/playback device to which the present invention is applied.

The audio recording and playback device 10 includes a central processing unit (CPU) 11
．． It has a memory circuit (A) 12A, a memory circuit (B) 12B, a speech synthesis circuit (A) 13A, and a speech synthesis circuit (B) 13B.

Both storage circuits 12A and 12B are comprised of RAM type memory LSIs and store audio information as digital signals.

Both voice synthesis circuits 13A and 13B are composed of voice synthesis LSIs, and each memory circuit 12A converts voice information into a digital signal.
, 12B and record data, and each memory circuit 12
In order to read the audio information written in A, 12B and reproduce the audio information as an analog signal, the control unit 14, A
D converter 15, I) A converter 16. A common microphone 19 is connected via an input amplifier 18 to each AD conversion section 15 of the two-way speech synthesis circuits 13A and 13B equipped with a switching section 17. The speaker (1) 21A is connected to the DA converter 16 of the speech synthesis circuit 13A via the output amplifier 20A by the switching operation of the switching unit 17, or the speaker (1) 21A is connected to the output amplifier 20.
A speaker (2) 21B is connected via B. Also,
The DA conversion section 16 of the speech synthesis circuit 13B includes a switching section 17.
By this switching operation, the speaker (3) 21C is connected via the output amplifier 20C, or the speaker (4) 210 is connected via the output amplifier 20D. In the present invention, the means for outputting audio information is not limited to the speaker itself, but also refers to the terminal section itself to which the speaker is connected.

Each speech synthesis circuit 13A, 13B processes two pieces of speech information (At, A2) (Bl, B2), respectively.

In the recording normal mode, the storage circuit 12A stores the two audio information (Al, A2) processed by the audio synthesis circuit 13A as digital signals at different addresses, and the storage circuit 12B stores the two audio information (Al, A2) processed by the audio synthesis circuit 13B as digital signals. Audio information (Bl, B2) is stored as a digital signal at different addresses. That is, the audio recording and reproducing device 10
As shown in the figure, four pieces of audio information (AI, A2, B1,
It is equipped with four storage systems, channels 1 to 4, each of which can store channels B2). Here, speech synthesis circuit 1
Each channel grouped by 3A and channel 2
The group of channels 3 and 4 combined by the voice synthesis circuit 13B will be called sector B.

In recording expansion mode, storage circuits 12A and 12
Both of the voice information processed by the voice synthesis circuit 13A (
A1) is stored as a digital signal.

The central processing unit 11 includes a recording mode setting section 22, and the recording mode setting section 22 includes a recording mode setting switch 23,
Recording time setting switch 24, recording start switch 25
, a recording stop switch 26.

■When you press the recording mode setting switch 23 once, the lamp 2
Only channel 2A is lit, and the recording normal mode (2) is set in which recording can be performed on channel 1 or channel 2 within all addresses of the memory circuit 12A.

A channel is selected by pressing the channel selection switch 27. At this time, the selected channel (l or 2) lights up on the channel display section 28.

■Press the recording mode setting switch 23 twice, the lamp 2
Only 2B is lit, and the recording normal mode (2) is set in which recording can be performed on channel 3 or channel 4 within all addresses of the memory circuit 12B.

A channel is selected by pressing the channel selection switch 27. At this time, the selected channel (3 or 4) lights up on the channel display section 28.

(■ When you press the recording mode setting switch 23 three times, both the lamps 22A and 22B turn on, and the memory circuits 12A and 12
A recording expansion mode is set in which recording can be performed only on channel l within all addresses of both channels. At this time, the channel selection switch 27 selects the channel 1, and the channel display section 28 lights up the channel 1.

The recording time in each recording mode described above is set by the recording time setting switch 24. The set recording time is displayed on the time display section 29. Here, central processing unit 1
1, in the above-mentioned recording normal mode (■), (■), Q) write addresses of the audio information (Al) and audio information (A2) to the storage circuit 12A corresponding to the audio synthesis circuit 13A,
The memory circuit 12 is mutually expandable within the range of all addresses of the memory circuit 12A, and corresponds to the voice synthesis circuit 13B.
The write addresses of audio information (Bl) and audio information (B2) for B can be mutually expanded within the range of all addresses of the storage circuit 12B. As a result, when the recording time is set to, for example, 60 seconds, it is possible to record 30 seconds each in recording normal mode ■ and ■, and 30 seconds.
If it is within seconds, the time allocation for each channel in that sector is free (e.g. 25 seconds for channel l, 5 seconds for channel 2, etc.). 60 seconds of recording is possible.

The central processing unit 111 controls the recording mode setting switch 2.
3. When the recording start switch 25 is pressed with the recording mode, recording channel, and Q tone time set by the recording time setting switch 24, the control section of the voice synthesis circuits 13A and 13B corresponding to the selected channel 14
Give the command to start recording to. The control unit 14 includes the memory circuit 12
The corresponding addresses of A and 12B are set to write mode, and the audio information injected into the microphone 19 is converted into a digital signal by the AD converter 15, and the digital signal is written to the corresponding address of the memory circuits 12A and 12B.

At this point, at the same time as the recording start switch 25 is pressed, the display time on the time display section 29 increments every second to display the remaining recording time.

To stop recording within the set recording time, press the recording stop switch 26.

The central processing unit 11 includes a playback mode setting section 30, and the playback mode setting section 30 includes a playback mode setting switch 31,
Playback interval setter 32, endless switch 33
, and a playback start switch 34.

By pressing the reproduction mode setting switch 31 and the channel selection switch 27, a channel to be reproduced is selected, and the selected channel lights up on the channel display section 28. It is also possible to turn on all channels 1 to 4 at the same time. Here, sector A
Since any of the channels within sector B and any of the channels within sector B are controlled by different audio synthesis circuits 13A and L3B, audio information can be reproduced at the same time.

On the other hand, channels within the same sector cannot be played simultaneously.

The start of reproduction is started based on the activation timing detected by sensors (1) 35A to (4) 35D corresponding to channels 1 to 4, respectively.

Here, the central processing unit 11 sets the playback interval by operating the channel selection switch 27 and the playback interval setter 32, thereby determining the activation detected by each sensor 35A to 35D within the playback interval time. Ignoring the timing, the speech synthesis circuit 13A,
13B may not be used for starting (reproducing) 13B. That is, as shown in the flowchart of FIG. 3, the central processing unit 11 adopts the activation timing detected by each sensor 35A to 35D, and then uses the activation timing newly detected by each sensor 35A to 35D after the regeneration interval time has elapsed. The activation timing is adopted as the timing for the next activation of the speech synthesis circuits 13A and 13B.

In the playback interval setting, a predetermined playback interval time is set in the playback interval setter 32 while a predetermined channel is selected by the channel selection switch 27.

The set reproduction interval time is displayed on the time display section 29.

It is also possible to start playback manually. In this manual start method, reproduction is started by turning on the reproduction start switch 34. At this time,
When the endless switch 33 is turned on, endless playback occurs, and when a predetermined playback interval time is set in the playback interval setting device 32, playback is performed repeatedly at intervals of the playback interval time.

Here, in the endless setting, each time the endless switch 33 is pressed, the lamps 30A and 30B are turned on in sequence, and when only the lamp 30A is turned on, only sector A becomes endless playback, and only the lamp 30B is turned on. When the lamps 3OA and 30B are turned on, only sector B undergoes endless reproduction, and when both lamps 3OA and 30B are illuminated, both sectors A and B undergo endless reproduction.

The central processing unit 11 controls the playback mode setting switch 31.
．． Playback interval setter 32, endless switch 3
3, when the sensors 35A to 35D detect the start timing or the playback start switch 34 is pressed,
a speech synthesis circuit 13A corresponding to the selected channel;
A reproduction start command is transmitted to the control unit 14 of 13B. Control part 1
4 sets the corresponding addresses of the storage circuits 12A and 12B to read mode, reads out the audio information stored in the storage circuits 12A and 12B, and transfers the audio information to the DA converter 16.
The signal is converted into an analog signal by the FJJ converter 17, and outputted from the corresponding speakers 21A to 21D.

Note that when both channels 1 and 2, or both channels 3 and 4 are to be reproduced, they are alternately reproduced in endless reproduction.

Also, when the playback interval time of channel 2 is reached, if the playback of channel 1 is not completed,
Channel 2 automatically goes into standby mode, and channel 1
Playback starts directly from the end of the playback.

In addition, in the above embodiment, each sensor 35A to 35D,
Each of the speakers 21A to 21D is arranged at a sound generation position such as a different sales floor. As a result, each sensor 3 detects objects to be sensed, such as people who are generating sounds, such as customers who visit each sales floor.
5A to 35D, and as a result, audio information such as advertising messages specific to each sales floor is transmitted to each speaker 21A to 35D.
21D to the person to whom the voice is generated.

Next, the operation of the above embodiment will be explained.

According to the above embodiment, the voice synthesis circuits 13A and 13B are activated at the activation timing detected by the sensors 35A to 35D, and the voice information recorded in the storage circuits 12A and 12B is transmitted through the voice synthesis circuits L3A and 13B. Speaker 21A
210, all activation timings newly detected by the sensors 35A to 35D within the reproduction interval time are ignored. That is, the next activation of the voice synthesis circuits 13A and 13B is performed at the activation timing newly detected by the sensors 35A to 35D after the reproduction interval time has elapsed. Therefore, even if the object to be sensed, such as the person to whom the sound is generated (such as a customer visiting the sales floor), activates the sensors 35A to 35D frequently,
Since the audio information is not outputted repeatedly, and the first output of audio information is based on the detection results of the sensors 35A to 35D, the audio information can be detected at appropriate timing with unexpectedness. It will be supplied to the following persons.

Note that the following modifications can be made in implementing the present invention.

■The audio generator may have only a playback function; in this case, it includes a recording mode setting section, a microphone, an AD
A ROM type memory LSI or the like may be used as the semiconductor memory element constituting the memory circuit by removing elements for recording such as a converter.

(2) The ROM type memory LSI constituting the storage circuit of the audio generator may be of a cassette exchange type.

[Effects of the Invention] As described in 2 below, the present invention includes a storage circuit that stores audio information as a digital signal, and reads audio information written in the storage circuit and reproduces the audio information as an analog signal. In a speech generating device comprising a speech synthesis circuit, a sensor for detecting the activation timing of the speech synthesis circuit, and a means for outputting speech information of the speech synthesis circuit, a predetermined reproduction interval time is set, and the sensor detects the activation timing of the speech synthesis circuit. The apparatus includes means for ignoring activation timing detected within the reproduction interval time and not using it for activation of the speech synthesis circuit. Therefore, voice information can be supplied to the person to whom the voice is generated at an appropriate timing.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an example of an audio recording/playback device to which the present invention is applied, Fig. 2 is a schematic diagram showing four storage systems of the audio recording/playback device, and Fig. 3 is a state of interval playback using the sensor start method. FIG. 10...Speech recording/playback device, 12A, 12B...Storage circuit, 13A, 13B...Speech synthesis circuit, 14...
-Control unit, 16...DA conversion unit, 21A to 210...
・Speaker, 30... Playback mode setting section, 32...
Regeneration interval setter, 35A to 35D...sensor. Agent Patent Attorney Osamu Shiokawa Figure 2

Claims (1)

[Claims] (1) A memory circuit that stores audio information as a digital signal; a speech synthesis circuit that reads audio information written in the storage circuit and reproduces the audio information as an analog signal;
In a sound generation device that includes a sensor that detects the activation timing of a speech synthesis circuit and a means for outputting speech information of the speech synthesis circuit, a predetermined reproduction interval time is set, and the sensor detects the activation timing within the reproduction interval time. A voice generating device characterized by comprising means for ignoring detected activation timing and not using it for activating a voice synthesis circuit.