JPS61249116A - Voice output unit for programmable controller - Google Patents

Abstract

PURPOSE:To generate voice signals successively in an optionally designated link order by providing a means which generates a voice generating command at each time of the end of voice generation successively in the detected link order in response to detection of presence of an output start command. CONSTITUTION:A microprocessor (MPU) 501 takes in data from a system bus DB and decodes it as a voice output instruction to detect presence/absence of the output start command. Since the first message number, the second mes sage number, and the third message number are stored in bits 0-7, bits 24-31, and bits 16-23 as binary codes and a message link number 3 is stored in bits 12 and 13 as a binary code by a control system program corresponding to the voice output instruction in a system program ROM 502, the MPU 501 issues the voice generating command, where three messages are linked at intervals of one secon in accordance with this program in response to presence of the output start command.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of the Invention) The present invention relates to a novel output unit used in a converting block type programmable controller, and in particular to a new output unit capable of generating various audio signals according to the contents of a user program. Concerning the crab knit that was made.

(Summary of the Invention) The present invention enables generation of an audio signal with contents arbitrarily programmed by the user from the output unit itself, and also enables the generation of audio signals corresponding to a plurality of audio contents arbitrarily specified by the user. This allows them to occur sequentially with the specified connection order.

(Prior Art and its Problems) When controlling various industrial machines using a programmable controller, it is sometimes preferable to use voice in order to facilitate smooth interaction between the machine and a human.

In other words, if voice is used for dialogue between machines and humans, content can be conveyed more accurately compared to visual dialogue using a display board, and unlike visual dialogue using a display board, it is possible to convey the content more accurately than with visual dialogue using a display board. There is no need for monitoring, and therefore, as long as the operator is within sound range, the operator can be engaged in other work, which is preferable from the viewpoint of labor saving.

Furthermore, despite the general advantages mentioned above, there are control fields in industry that do not necessarily favor automation. for example,
In technical fields that require extreme reliability, such as nuclear equipment, there are cases in which administrative guidance requires manual operation of final valve operations, etc., due to fear of malfunction due to automation. .

In such a case, reliability can be further improved by broadcasting the manual operation procedure by voice.

To give a rough explanation using a familiar example, automatic valves such as solenoid valves are generally always provided with a bypass path, and a manual valve is inserted into this to ensure that the flow is not interrupted even when the solenoid valve malfunctions. However, even if it is possible to detect a malfunction of this automatic valve using a flow switch, etc., unless the operator confirms in advance which manual valve to open, it will not be possible to immediately It is not possible to respond to

In such cases, the operator (if it is possible to immediately notify the operator (or
It is possible to shorten failure recovery time and minimize damage to equipment.

As described above, there are various demands for the use of voice in dialogue between machines and humans in the industrial world, but in order to meet these demands using conventional programmable controllers, it has been necessary to use the technology since the beginning of programmable controllers. An endless tape device, which is installed separately from the output unit, can output binary numbers output with contact or non-contact.
A complicated and expensive configuration such as leading to an audio signal generating device such as an audio synthesis LSI had to be adopted.

(Objective of the Invention) The object of the present invention is to enable generation of audio signals with contents arbitrarily programmed by the user from the output unit itself, and to generate audio signals corresponding to a plurality of audio contents arbitrarily specified by the user. The purpose is to enable sequential generation using a specified connection order.

(Structure and Effects of the Invention) In order to achieve the above object, the present invention includes: a data importing means for importing data sent to the unit onto a system bus of a programmable controller; a command decoding means for decoding data as a voice output command and detecting each voice content to be concatenated and output, the concatenation order, and the presence or absence of an output start command; in response to the detection that the output start command is present; A sound generation command means that sequentially issues a sound production command related to each detected sound content in the detected concatenation order each time a sound is completed; and a sound signal generation means that generates a corresponding sound signal in response to the sound sound command. : It is characterized by comprising the following.

According to such a configuration, it is possible to generate an audio signal with content arbitrarily specified by the user from the output unit itself, and the audio signals corresponding to the arbitrarily specified audio content can be sequentially generated in the specified connection order. It can be made possible.

(Description of Embodiments) FIG. 1 shows the overall appearance of a programmable controller to which the present invention is applied.

As shown in the figure, this programmable controller 1
is power supply unit 2. The CPU unit 3 is composed of one or more I10 units 4 and a sound output unit 5 according to the present invention.

Each of the units 2 to 5 is configured with a desired circuit board built into a book case-shaped housing, and a plug protruding from the back of each housing is inserted into a socket on the system bus installed in the rack 6. By being plugged in, the units 2 to 5 are electrically connected to each other via the system bus.

Further, as will be described later, the sound output unit 5 generates a sound signal designated by the user program, and by amplifying this signal with the amplifier 7, the sound is produced from the speaker 8.

The power supply unit 2 rectifies and smoothes the commercial power, stabilizes it, and supplies it to the units 3, 4, and 5. The power supply unit 2 can turn on and off the power of the entire controller.

The I10 unit 4 is already known and will not be described in detail, but each channel has 16 points and can be assigned to one or more channels, and each point has a pair of inputs and outputs. The circuit is equipped with a built-in digital switch that allows 16
Can be freely set as input or output on a point-by-point basis.

Furthermore, in this embodiment, each unit 4 has a built-in digital switch for address setting, and by setting this appropriately, the socket position on the rack 6 is not restricted and any address setting can be made. A so-called free location method is adopted.

As shown in FIG. 2, the CPLJ unit 3 is an 8-bit microprocessor (MPU) 3 typically such as
0 as the main component, and its data bus DB,
The address bus AS and the control bus CB include an input/output state RAM 31 and a working RAM 32 . System program ROM33aJ: (In addition to connecting the F-L-f program PRoM34, socket 3 is used to connect a programming tool when writing a program, etc.)
5 is connected, and plug 36. It can be connected to the system bus SB on the rack 6 via the socket 37.

The input/output status RAM 31 has one word of 8 bits R.
AM is used, and the address space of this RAM 31 includes input/output channels that can be assigned as inputs or outputs in 8-bit units, as well as an auxiliary relay area, a count information area for counter elements, a keep relay area, and a timer area for timer elements. It is assigned to the information area, etc.

Similarly, the working RAM 32 is a 19-bit 8-bit RA.
M is used as a temporary storage area during various calculations in the MPU 30.

The system program ROM 33 is similarly composed of a memory of 8 bits per word, and stores various system programs necessary for the operation of the programmable controller. Details of this system program will be explained later.

Similarly, the user program PROM 34 uses a FROM of 8 bits per word, and this FROM 34 is programmed with the programmable controller language (high-level language).
A user program arbitrarily set by the user is stored. In this embodiment, the user program is stored in a format for inputting a ladder diagram.

FIG. 3 shows an example of a system program (in the case of the END refresh method) stored in the system program ROM 33. As shown in the figure, when the system program is started upon power-on, the input/output state RAM 31. The contents of the working RAM 32 and the like are initialized (step 300), and then various system service processes are performed (step 301).

In this service processing, in addition to service processing such as monitoring and failure diagnosis, if a programming tool is connected, processing for writing a program into the user program PROM 34 is performed.

When the service processing is completed, input update processing is subsequently performed (step 302). In this input update process, I
Among the 10 units 4, 4, . . . , a process is performed in which the states of the terminals assigned to the inputs are sequentially taken in in 8-bit units and transferred to the corresponding addresses in the input/output state RAM 31.

Thereafter, in the service process 301, unless a key operation corresponding to the RtJN mode is performed (step 303
negation), and the above operations are repeated (step 30).
1,302>, only input updates are performed continuously.

In this state, when a predetermined key operation corresponding to RUN mode is performed in service processing 301, RLIN mode is determined following input update processing (Yes at step 303), and command execution processing is performed (step 303). 304).

In this instruction execution process, user instructions are sequentially read from the user program FROM 34 according to the contents of the program counter, circuit calculations etc. are performed according to the contact connection relationships on the ladder diagram, and along with the execution of the OUT instruction, the calculation results up to that point are A process of rewriting the corresponding output canister in the input/output status RAM 31 is performed. When the END instruction is read while repeating instruction execution, the instruction execution process is ended and the subsequent output update process (step 305) is performed.

In this output update process, the contents of the output canister of the input/output state RAM, which have been rewritten as a result of instruction execution, are transferred to each of the eight terminals assigned as outputs of the I10 unit.

Thereafter, as long as the RUN mode continues, system service processing (step 301) and input update processing (step 302) are performed.
, instruction execution processing (step 304), and output update processing (step 305) are repeatedly executed.

In this embodiment, since the user program is written in the programmable controller language, it cannot be directly executed by the microprocessor 30, and therefore is executed via an interpreter program stored in the system program ROM. Executed indirectly.

Next, this kind of ladder diagram input type programmable
Among the OP codes used in the controller, OUT and MOV, which are necessary for understanding the sound output crab unit of this embodiment, are
The instruction execution operation will be explained.゛The ladder diagram shown in Fig. 4 (a) is expressed as shown in Fig. 4 (b) using the OP codes LD and 0tJT, and the instruction decoding results are shown in the flow chart in Fig. 4 (C). expressed.

That is, if the result of reading the instruction according to the program counter (step 400) is determined to be an OUT instruction (step 401 affirmative), it is determined whether the calculation result up to that point, that is, input 3000, is ON or not. (Yes at step 402), as shown in FIG.
1″ corresponding to N is entered, and the other input 3000 is OF
In the case of F (step 402 negative>), an OIU corresponding to OFF is written (step 404).

In this way, when the OP code OUT is executed, the corresponding output is always rewritten to the content corresponding to the calculation result up to that point.

Next, the ladder diagram shown in Fig. 5(a) is expressed as shown in Fig. 5(b) using the OP codes LD and MOV, and the instruction decoding result is shown in Fig. 5(C). Represented as a flowchart.

That is, when the result of reading an instruction according to the program counter (step 500) is determined to be MOV (step 501 affirmative), it is determined whether the calculation result up to that point, that is, the input 3000 is ON, and it is determined that it is ON. Only when it is determined (step 502 #determined), the data 1234 represented by the BCD code is stored in the input/output state RAM3.
1 O channel.

In this way, when executing MOV, which is an OP code, only if the calculation result up to that point is ON, the OP code is
The data of the BCD code written following the code MOV is stored in the input/output state RAM 31 as shown in FIG.
transferred to the channel.

In this way, unlike the case of the OUT code, in the case of the MOV code, the contents of the specified channel are rewritten with the data written in the program only when the previous calculation result is ON. In the case of OFF, no rewriting operation is performed.

The OUT and MOV codes described above are used when programming the audio output crab unit according to the present invention.

Next, the configuration of the voice output crab knit according to the present invention will be explained. As shown in Fig. 6, the five housings of the sound output crab unit 5 are formed in a book case shape similar to the unit 10, and the front panel has 32 housings for displaying the message number being sounded. LED indicator 52.
Jack 53 for taking out audio signals. 8-bit DIP switch 5 used for various operation settings described later
4 and a removable front cover 55.

By inserting a plug 56 into the jack 53, an audio signal is extracted and guided to an amplifier or monitor speaker (not shown).

Further, a space (not shown) for storing memory is provided behind the front cover 55, and by inserting a voice memory card 57 into this space, messages to be pronounced can be exchanged in units of 32 types. It is now possible.

Further, as described later, the voice memory card 57 is provided with an identification bit whose state can be set by attaching/detaching a jumper wire. However, it is possible to identify whether the instruction is for a bit-compatible instruction or a code-compatible instruction.

Furthermore, a plug 58 (not shown) is formed protruding from the rear surface of the housing 51, and this plug 58 is connected to the rack 6.
The CPU unit 3 is connected to the CPU unit 3 by inserting it into a socket 59 on the system bus installed above.

The electrical configuration of the audio output unit 5 is shown in FIG. As shown in the figure, the audio output unit 5 is mainly composed of an 8-bit microprocessor (MPU) 501, and its data bus DB'', address bus AB-
, control bus CB- has the system program R
A working RAM 503 is connected to the OM502.

A system program necessary for controlling the voice output unit is stored in advance in the system blog ROM 502, and the contents of this program will be explained in detail with reference to a flowchart to be described later.

In the working RAM 503, as shown in FIG. A second stack area is provided, and as described later, these stack areas contain the 32-bit audio output instruction itself in the case of a code-compatible instruction, and the decoding result in the case of a bit-compatible instruction. Equivalent BCD
The message number represented by the code is stored. Note that the functions of these stack areas will be explained in detail later.

The data bus DB in the audio output unit 5 and the data bus DB in the programmable controller are the socket 59, the plug 58, the latch section 504, and the latch section 50.
5.

The latch unit 504 has four 8-bit latch circuits connected in parallel, and each latch circuit is selectively accessed via an address decoder.

Therefore, from the CPU unit 3 to the address bus DB,
The data transmitted in a time-division manner in four 8-bit units is converted into 32-bit data in parallel by the action of the latch section 504. Furthermore, particularly in this example, the 14th bit after parallel conversion is supplied to the MPU 501 as a dedicated interrupt signal (interrupt 2).

The latch unit 505 has four 8-bit latch circuits arranged in parallel, and the outputs of the gates provided on the output side are commonly connected in units of 8 bits, so that each latch circuit is triggered simultaneously, and each gate is By selectively operating the address decoder output, parallel 32-bit data,
In other words, audio output commands are sent to the data bus D in units of 8 pits.
It is possible to send it out onto "B".

The LED display 52 has a 32-bit configuration as described above, and is arranged on the upper part of the front panel of the housing 51. This LED indicator 52 is connected to the latch section 5 described above.
It is connected to the data bus DB- through a latch section 506 having substantially the same configuration as 04.

The DIP switch 54 has an 8-bit configuration as described above, and is mounted on the front panel of the housing 51. This DIP switch 54 has a gate section 507
The data bus DB- is connected to the data bus DB-.

Next, the configuration of the audio signal generator will be explained. In this embodiment, the ADPCM method (Adapt
ive Differential Pulse Co
ded hoduration) has been adopted, thereby improving the sound quality. Note that the configuration of the audio signal generating section may be a PARCOR system or the like.

The audio data ROM 508 is the audio memory card 5 described above.
7, and is composed of a memory of 8 bits per word as shown in FIG.

This audio data ROM 508 is provided with an address storage area 508a in which the message number, a separate start address, and a stop address are stored, and an audio data storage area 508b in which the message number and audio data are stored separately. .

In addition, a jumper wire 509 is connected to the audio data ROM 50B.
An identification bit whose state can be set by connecting and separating is provided, and the contents of this identification bit can be read onto the data bus DB= in response to a specific address. Depending on the contents of this identification bit, it is possible to identify whether the instruction is for a bit-compatible instruction or a code-compatible instruction.

The voice synthesis unit 510 synthesizes a minute time worth of voice signals in response to 1 byte of data output from the voice data ROM 508, and sends an interrupt signal (interrupt 1) back to the MPU 501 every time the synthesis is completed. It looks like this.

The low-pass filter 511 removes harmonic components from the audio signal output from the audio synthesizer 510. Then,
The audio signal from which harmonic components have been removed is branched into two systems, one system being supplied to the primary side of the transformer 512, and the other system being supplied to the input of the monitor amplifier 513.

The secondary output of the transformer 512 and the output of the amplifier 513 are selectively connected to the jack 53 via a changeover switch 514.
This makes it possible to derive the changeover switch 5.
14 to the amplifier i13 side, it is possible to connect a monitor speaker directly to the jack 53 and perform audio tests, etc. Note that this changeover switch 51
4 is attached inside the housing 51, and is operated by removing the lid 55.

Furthermore, since the output of the low-pass filter 511 is branched into two systems, one of which is led out to the jack 53 via the transformer 512, and the other is led out to the jack 53 via the monitor amplifier 513, In addition to the basic effect that the amplified audio signal and the original audio signal can be selectively routed to the same jack, even when the original audio signal is routed to an external amplifier, the external amplifier side Since signals can be extracted with the ground as a reference, the ground potential is stable and less susceptible to noise and the like.

That is, if the output of the low-pass filter 511 is led directly to the jack without going through the transformer 512, it is necessary to connect the ground of the external amplifier and the ground of the audio crab unit 5 in common. Although a potential difference occurs between the two and the signal tends to be affected by noise, the present invention does not require such considerations.

Next, the configuration of the control system program stored in the system program ROM 502 will be explained.

This control system program is compatible with two types of audio output commands having different command formats.

FIG. 10 shows the audio output command in the first format.

This audio output command consists of 32 bits of O pitles.
It is composed of bit length data, and each content of the 31 bits of O pits corresponds to message No. O to message No. 31, respectively. For example, the 7th bit is "'
1'' means that the message content to be sounded corresponds to message No. 0.7.

Further, the content of each bit changes from "0" to "1" to indicate that there is an instruction to start audio output of the message corresponding to the bit. Hereinafter, this first format audio output command will be referred to as a pit compatible command.

The audio output command in the second format is shown in FIG. This audio output command also consists of 32-bit length data consisting of 31 bits of O pitles.

Then, a 1-byte portion consisting of 0 bits to 7 bits, 2
1 byte part consisting of 4 bits to 31 bits and 16 bits
The 1-byte portion consisting of bits to 23 bits contains message numbers corresponding to the contents to be pronounced first, second, and third, respectively.

is stored in BCD code.

In addition, the 4-bit part consisting of 8 bits to 11 bits stores the number of repetitions, and the 12 bits to 13 bits store the number of message concatenations.The 14th bit is used as a message recognition bit, and is stored in "0 to 13 bits". In addition to indicating the presence of a voice output start command by changing the state to “1'',
The 15th bit is a priority function selection bit, and 1'' indicates that the priority function is selected. Hereinafter, this second format audio output command will be referred to as a code-compatible command.

Next, on the premise of each of the above instruction formats, the operation of the system program will be explained starting from the case of bit-compatible instructions.

Note that, as will be explained in detail later, pit-compatible commands are executed by the CPU.
In order to send data from the unit 3 to the audio output unit 5, for example, assign the address of the audio output unit to channels 0 to 1 (one channel is 16 bits) in the input/output address space, and use the OUT command in the user program. , the output relay corresponding to the desired message number may be programmed to operate at the desired timing.

If such programming is performed, the 32-bit data corresponding to the desired pit-compatible command will be transferred from the input/output state RAM 31 in the CPU unit 3 to the audio output crab unit 5 by the aforementioned output update process (step 305 in FIG. 3).
The data is automatically transferred to the latch unit 504 inside.

In Fig. 12, when the program is started by turning on the power, etc., the general well-known initialization process (
(not shown), a determination process is performed as to whether the instruction type is a pit-compatible instruction or a code-compatible instruction (step 120).
1>.

This determination process is performed by reading the state of the identification bit set by the jumper line 509 of the audio data ROM 508 via the data bus DB-.

Here, if it is determined that it is a pit compatible command (step 12
01 bit corresponding command), the 1st . After initializing the second stack area and the like (step 1202), a series of operations from command decoding to generation command is started.

Suppose now that all 32 bits of output relays assigned to the sound output unit are OFF, the sounding flag F1 (described later) is set to the non-sounding state, and the first... Second
The stack area is in an empty state, and the rough test switch (corresponding to the second bit of the DIR switch 54) is OFF.
Assume that N.

In this state, when the program is started, if there is no BCD code corresponding to the message number in the stack and there is no BCD code corresponding to the message number in the stack,
4 negative) and the test switch 0FF (step 1205 negative), the latch unit 50
The 32-bit length pit corresponding instruction latched in bit 4 is transferred and latched in parallel to the latch unit 505.

Next, this latched pit corresponding instruction is read in a time-sharing manner four times each with 8 bits, and is compared with the previous read data to obtain "0" (step 120).
7 (No), the process returns to step 1203 and the above operations are repeated.

On the other hand, when any one of the output relays assigned to the audio output unit υj turns ON as a result of the command execution in the CPU unit 3, it is determined that there is a change bit, that is, there is an audio output start command (step 1207). (step 12), then instruction decoding processing is performed (step 12).
08>.

In this instruction decoding process, it is detected which of the 31 O pitles the bit that changed from "0+e" to "1°" is, and the detected bits 1 to 1 are assigned to the message No.
, and the corresponding BCD code is stored in the first stack area in the working RAM 503 shown in FIG. Note that, as will be described later, in the case of bit-compatible instructions, the first
．． There is no functional difference in the second stack area.

In the next operation cycle starting at step 1203, it is determined that there is a BCD code corresponding to the message No. in the stack (step 12041), and after passing through pause processing for one second (step 1209), the process shifts to sound production command processing.

Note that this one-second pause process (step 1209) is
As will be described later, this is for inserting a silent period of one second between messages when a plurality of messages are to be sounded continuously.

There are two types of sound generation command processing: with and without a priority function, and these are determined by the priority function selection switch (DIP).
It is set by reading the state of the bit (corresponding to the Oth bit of the switch 54).

Here, the case without the priority function will be explained first, and the case with the priority function will be explained in detail later.

If it is determined that there is no priority function based on the state of the priority function selection switch (No in step 1210), the first. Second
The message number stored first among one or more message numbers stored in the stack area,
is read out (step 1211), and the voice data ROM 50.8 is read out according to the read message number.
is accessed, and the start and stop addresses corresponding to message NO0 are read (step 1213). In addition, at this time, 1. The second stack area operates as a serial FIFO stack, and read message numbers and data are lost.

Next, the address of the audio data ROM 508 is set based on the read start address (step 1214), and the message N is set via the latch section 506.
After the LED corresponding to o is turned on (step 1215>), a start command is finally issued to the speech synthesis unit 510, and at the same time, the sounding flag F1 is set to "sounding" (step 1215). 1216>. After that, the process returns to step 1203 and moves to the next execution cycle.

On the other hand, on the audio data ROM 508 side, according to the setting of the start address, 1 byte of data stored at the corresponding address is sent to the audio synthesis section 510. In response to the start command, the speech synthesis section 510 performs speech synthesis processing, synthesizes a minute period of speech signals corresponding to the corresponding message number, and returns an interrupt signal (interruption 1) to the MPU 501 when the synthesis is completed. do.

Then, on the MPU501 side, the interrupt signal (interrupt 1)
In response to this, the sound generation continuation process shown in FIG. 13 is executed thereafter.

That is, first, after confirming that there is no interruption command (described later) (step 1301, negative), in the case of a bit-compatible instruction (step 1302, bit-compatible), the command is executed until the current set address reaches the above-mentioned stop address (step 13).
03 (No), and each time an interrupt signal @ (interrupt 1) is returned from the voice synthesis section 510, the currently set address is sequentially incremented (step 1304), and the voice data ROM 508 is set to the new address (step 1305).

On the other hand, if the current set address has reached the stop address (step 1303 affirmative), a stop command is issued to the speech synthesis unit 510, and the sounding flag F+ is set to "non-sounding".

(Step 1306), and at the same time, the LED corresponding to the message NO., which had been sounded until then, is turned off (Step 1307).

Thereafter, when the repetition number information given by the third to sixth bits of the DIP switch 54 is rOJ, (
Step 1308 (Yes), immediately ends the pronunciation continuation process, completes the pronunciation process for one round of the voice related to the message number, and waits for the next pronunciation command.

Furthermore, if the repetition information read from the third to sixth bits of the DIP switch 54 is rNJ, a sound generation command is issued for the same message number N+1 times (steps 1309, 1310, 1311> .

When the above series of processing is performed, the speech synthesis unit 510 outputs an audio signal corresponding to the message No.
This audio signal passes through a low-pass filter 511 to remove harmonic components, and then goes directly to the jack 53 via a transformer 512 or to the jack 53 via a monitor amplifier 513, depending on the setting state of the changeover switch 514. is supplied alternatively.

Therefore, the original audio signal can be extracted from the audio crab unit 5 itself by simply inserting the plug 56 into the jack 53 as shown in FIG. As long as you connect an amplifier to the speaker, you can hear the corresponding message from the speaker.

Furthermore, by inserting the plug 56 into the jack 53 with the selector switch 514 connected to the amplifier 513 side, it is possible to obtain a somewhat amplified audio signal, which is then directly supplied to the monitor speaker. You can perform an audio output test.

As a result of decoding the bit-compatible audio output command, if the presence of an audio output start command is detected and no audio signal is generated at the time of detection, the content related to the detected message No. 1 is audio signals can be generated immediately.

Furthermore, if repeat count information N is given from the third to sixth bits of the DIR switch 54 at this time, the same message can be repeatedly generated N times with a silent period of one second.

Note that when decoding a bit-compatible instruction, if two or more output relays are turned on at the same time during the same instruction execution cycle on the CPU unit side, among the turned-on relays,
The message with the smallest message number is given priority.

Next, the process to be performed when an audio signal corresponding to one of the messages is already being generated at the time when the presence of an audio output start command is detected is divided into the case without the priority function and the case with the priority function. I will explain.

First, to explain the case without the priority function, in FIG. 12, the presence of an audio output start command is detected as a result of decoding the audio output command obtained during the generation of some audio signal (step 1203). (step 1218, affirmative), and after confirming that there is no priority function (step 1219, negative), the detected message number is pushed to the first and second stack areas.Here, if there is no priority function, the first and second stack areas. The second stack area functions as a serial FIFO stack.

Thereafter, until the currently generated audio signal ends (step 1203 affirmative), the system waits while repeating the acquisition of audio output commands (step 1217) and decoding, and during this time, the presence of an audio output start command is detected. If so (step 1211), it is confirmed that there is no priority function (step 1219 negative), and the detected message numbers and data are sequentially bushed into the stack area in the same manner as described above (step 1224).

During this time, if the audio signal that has been generated up to that point ends (step 1203: negative), it is determined to proceed to step 1204 to proceed to the above-mentioned sound generation start command.

1209>, after confirming that there is no priority function (step 1
210 negation), 1st. The second stack area is connected to the F in series.
It functions as an IFO (First In First Out staff'), successively stores the oldest message number stored first, and performs the above-mentioned pronunciation command processing (steps 1213 and 1214) for this message number.
．． 1215, 1216> and the sound generation continuation process shown in FIG. 13 is executed.

Thereafter, each time the generation of the audio signal for the message ends (No in step 1203), a newer message NO.
, are sequentially read out (step 1211), and similar sound generation command processing and sound generation continuation processing are executed.

In this way, if one or more other audio output start commands arrive during some audio signal generation time (up to 30 seconds in this example) and there is no priority function, in this embodiment 1 in sequence every time the audio signal ends.
It is possible to generate audio signals corresponding to each message number at intervals of seconds in the order in which they arrive.

Next, to explain the case where the priority function is present, as mentioned above, the presence of the voice output start command is detected (step 12).
18 (affirmative), and if it is determined that there is a priority function (
In step 1219 (Yes), a priority comparison is made between the message currently being sounded and the message that has arrived. In this example, it is decided that the smaller the message number, the higher the priority.

Here, if the incoming message has a lower priority (No in step 1220), the detected message No. is immediately pushed to the stack area as in the case without the priority function.

On the other hand, if it is determined that the incoming message has a higher priority (step 1220), a stop command is sent to the speech synthesis section 510 (step 1221), and at the same time message No. Corresponding LE
D is turned off (step 1222), and the message number and data currently being sounded are returned to the stack area.

As a result, the message currently being sounded is forcibly interrupted.

Thereafter, until the message currently being produced ends (step 1203 affirmative), the voice output command is repeatedly decoded (step 1217>, and if a new voice output start command is detected during that time (step 1218) (Yes), one of the processes according to the priority order described above is performed.

On the other hand, when the message that has been being sounded ends or is forcibly interrupted as described above (step 1203
After confirming that the message No. and data are present in the stack (step 1204 affirmative), the process proceeds to the above-mentioned sound generation start command.

In the sound generation command with priority function (Step 1
210 affirmative), the smallest message number at that time among the message numbers stored in the stack area,
is read out, and the above-mentioned sound generation command processing (steps 1213 and 1214) is performed according to the read message number.
．． 1215, 1216) and the sound generation continuation process shown in FIG. 13 is performed.

Then, each message number stored in the stack area
, each time the generation of the audio signal corresponding to , is completed (No in step 1203), the smallest message number excluding those that have already been uttered is sequentially read from the stack (step 1212>), and similarly, the generation command processing and the generation are performed. Continuation processing is performed.

In this way, with the priority function, if one or more production commands arrive while a certain message is being produced, among all the messages that are currently receiving production commands, including the message that is currently being produced, Most message No.
Messages with a smaller number are preferentially pronounced sequentially.

Therefore, the voice data RO is set so that the message number for a message with a higher degree of urgency becomes a smaller value.
By configuring M, it becomes possible to create an audio output program that takes priority into consideration.

Next, a pronunciation test will be described in which a specified message is produced by a command (simulated sound output command) from a switch provided on the sound crab knit 5 side.

In this case, the message number, ON No., or 31 is specified using 5-bit binary data consisting of the third to seventh bits of the DIP switch 54, and the test switch (corresponding to the second bit of the DIP switch 54) Set from OFF to ON.

Turn the test switch from OFF to ON while the program is running.
is detected (step 1205 affirmative), then DI
The message numbers and data of the 3rd to 7th bits of the P switch 54 are read (step 1225), and the above-mentioned sound generation command processing (steps 1213, 1214, 1) is then carried out.
215.1216> and the sound generation continuation process shown in FIG. 13 is executed.

Therefore, after setting the message number in binary code, by turning on the test switch, the audio signal corresponding to the message can be immediately generated.
By switching the selector switch 54 to the monitor amplifier 513 side and connecting a monitor speaker to the jack 53, the contents of the message can be confirmed at a relatively low volume. At the same time, the LE corresponding to the message number being sounded
By lighting up D, it is possible to easily confirm the relationship between the content of the message being pronounced and the message number.

Roughly speaking, if the test switch is turned from ON to OFF while some message is being generated, it will be recognized that a message interruption command has been issued during the generation continuation process (step 1301).
, a stop command is immediately issued to the speech synthesis unit 510 (step 1312), and the LED is also turned off (step 1313), so that the message can be pronounced at any time without having to listen to the entire message for up to 30 seconds. can be forcibly interrupted.

Therefore, when checking the contents of messages sequentially,
If you can roughly understand the content from just the initial part of each message, you can use this interrupt function to shorten the test time by listening to only the first part of each message. becomes.

As explained in detail above, the operations in response to bit-compatible commands include (1) immediately generating the corresponding voice in response to the detection of a voice output start command, and (2) the number of repetitions specified on the unit side. (3) During some kind of pronunciation,
When one or more other pronunciation commands arrive, the operation waits for the previous pronunciation to finish and then sequentially pronounces each message in the order in which they arrive; (4) Similarly, one or more other pronunciations are made during the pronunciation. When a command arrives, among all the messages that arrived after interrupting the sound production, the message number is given priority in descending order of number. (5) Responds to the simulated voice output command given from the unit side. do,
It is possible to produce or interrupt the sound of a specified message number.

Next, the operation of the system program corresponding to the code-compatible instructions will be explained.

In FIG. 12, when the power is turned on, the program is started, and when it is determined that it is a code-compatible command by referring to the identification bit of the audio data ROM 508, the program moves to the flowchart of FIG. 15, and the first . After initializing the second stack area, etc. (step 1501), a 4-byte audio output command is stored in the first stack area, provided that no message is being produced (step 1502, negative). (Yes at step 1503), the audio output command is stored in the second stack area (step 150)
41 constant) or the test switch (corresponding to the second bit of the DIR switch) is turned on (step 1505
(affirmative).

On the other hand, in the case of a code-compatible voice output command, the 14th bit is assigned as a message recognition bit, as shown in FIG.
A change from `` to 1'' indicates that an audio output start command has been issued.

Furthermore, as shown in FIG.
The bit-th output is M as an interrupt signal (interrupt 2).
It is supplied to PtJ501.

As will be explained briefly and in detail later, in order to give a code-compatible audio output command to the audio output unit 5, in order to give the audio output command corresponding to the code, the audio A program is set up to transfer 32-bit data corresponding to an output command using a MOV command, and to turn on output relay No. 14 every time the data is transferred.

Therefore, when the MOV instruction for transferring audio output commands and the 0UT14 instruction paired therewith are activated during execution of the user program, the 14th bit of the latch unit 504 in the audio ratio controller 5 is output. changes from “Oat” to “1”, and this state change causes the MPU50
1, and the data storage process shown in FIG. 14 is activated.

Then, the audio output command latched by the latch unit 504 is transferred to the latch unit 505 and latched (step 140).
1) The audio output command latched by the rough latch unit 505 is read into the MPU 501 via the data bus DB'' (step 1402>).

Thereafter, various audio output modes are determined depending on the content of the read audio output command, but first, as the simplest example, the audio with the content corresponding to a certain message number is output once. We will explain the case where only the following words are pronounced.

In this case, in the format shown in FIG. 11, the message number to be sounded is set to B bits
It is stored as a CD code, and the number of repetitions of 8 bits to 11 bits is rob, 12 bits.

The 13-bit message concatenation number is rlJ, and the 15th bit priority function selection bit is set to "Ope."

A voice output command with such content is read (step 1)
402>, and no message is being pronounced at this reading time (step 1403 negative), after confirming that there is no priority function ゛ (step 1404 negative)
, the audio output command is transferred to the second stack area for exclusive use without a priority function, which functions as a FIFO stack (step 1405).

Then, on the main flowchart side of FIG. 15, the presence of an audio output command is detected in the second stack area (
Step 1504 (Yes), then after passing through pause processing for one second (Step 1506), the first message number of the oldest voice output command (in this case, the message number that has just been stored) is read out from the second stack area ( Step 1507), from now on, the sound generation command processing for this message No. (Steps 1508, 1509, 1
510.1511) and the pronunciation continuation processing shown in FIG.
604.1605> is performed, and the message No.
The audio related to the content will be pronounced only once.

On the other hand, if one or more voice output start commands are detected during any voice pronunciation (step 1
403 affirmative), after confirming that there is no priority function (step 1
406 (No), and the read audio output commands are sequentially pushed to the second stack area (step 140).
7).

Then, on the main flowchart 1 side shown in FIG. 15, while it is determined that the message NO and data are present in the second stack area (step 15041), there is a pause of 1 second due to the FIFO stack function (step 1506). ), a pronunciation command is given to the speech synthesis unit 510 (steps 1508, 1509, 1510.1).
511>, the sound generation continuation process shown in FIG. 16 is further executed, and one or more new messages that arrive during the sounding of any message are successively sounded at 1-second intervals in the order in which they arrive. .

Next, if it becomes necessary to pronounce a message with a high degree of urgency while a message with a low degree of urgency is being pronounced, the pronunciation of the message with a low degree of urgency is interrupted and the message with a high degree of urgency is prioritized instead. The case of making sounds will be explained. In this case, in the instruction format shown in FIG. 11, the 15th priority function selection bit is set to "1" compared to the above case.

In FIG. 14, when a priority function is detected while any message is being sounded (step 1403 is affirmative),
Step 14064), it is further determined whether or not the message currently being generated has a priority function, and if the message being generated does not have a priority function (No in Step 1408), the voice output command currently being generated is If it has a stack function but does not have a priority function, it is returned to the dedicated second stack area (step 1409), and a stop command is sent to the speech synthesis unit 510 (step 14).
10) and turning off the corresponding message (step 14)
After passing through step 11), the newly obtained voice output command is pushed to the first stack area dedicated to the FIFO stacking function and the priority function (step 1412).

Then, in the main flowchart of Figure 15,
The presence of a voice output command is detected in the first stack area (step 1503 affirmative), and after passing through a pause process for 1 second (
Step 1512), the -1st message N01 (in this case, the message number just stored) of the oldest voice output command is read from the first stack area (Step 1513), and this read message N
Regarding O, the above-mentioned sound generation command processing (step 150
8°1509.1510.1511) and the sound generation continuation process shown in FIG. 16 is executed.

As a result, messages with low urgency related to non-priority functions that were being sounded are forcibly interrupted, and instead, messages with high urgency that have arrived during sounding are sounded preferentially.

In addition, if similar voice output commands related to the presence of the priority function arrive one after another while a message related to the presence of the priority function is being produced (step 1408 pollution), these voice output commands are sequentially transferred to the first stack area. (step 1412).

Then, as a result of executing the stack read processing (FIFO) on the main flowchart side (step 151
3) In the case of a message from a comrade with a second function, messages will be sounded in the order in which they arrive at one-second intervals.

The above is a summary of the processing when the presence of another voice output start command is detected during sound generation. During the sound generation of a voice output command related to no priority function, voice output commands for both those with priority function and those without priority function If multiple start commands related to the priority function are given, at the time when the start command of the audio output command related to the priority function is given, the pronunciation related to the non-priority function is forcibly interrupted, and then the sound related to the priority function is interrupted. Only the output commands are sounded at 1-second intervals in the order in which they arrive, and then the interrupted voice output commands related to the no-priority function are sounded again from the beginning, and when the sound generation ends, the voice output commands related to the no-priority function are Similarly, the sounds are made to sound sequentially at 1 second intervals in the order in which they arrive.

On the other hand, if multiple start commands related to priority function presence and non-priority function arrive during the sound generation related to priority function presence, the above-mentioned forced interruption processing of the voice is not performed, and the voice processing related to priority function presence is simply performed. After the voice output commands are sounded in the order in which they arrive, the voice output commands related to the non-priority function are similarly sounded in the order in which they arrive.

In this embodiment, the priority function selection bit is set to 1 bit and the priority level is set to 2 levels, but it goes without saying that this may be set as a plurality of bits and the priority level may be set in multiple stages.

Next, a case will be described in which three messages are concatenated in a specified order and pronounced.

In this case, in the instruction format shown in FIG. 11, bits 0 to 7 contain the first message NO.
For 24 bits to 31 bits, the second message number,
However, the third message No. is stored in the 16th to 23rd bits as an 8CD code.

Furthermore, a binary code corresponding to the number of message connections "3" is stored in 12 bits 1 to 13 bits.

For the sake of simplicity, let us assume that there is no priority function and that no message is being pronounced.

In such a state, when an audio output command is read in response to a predetermined start command (step 1402 in FIG. 14), the audio output command is pushed to the second stack area (step 1403 negative, 1404 negative,
1405), then the series of processes in FIG. 15 (step 1
504 to 1511) and the interrupt processing shown in FIG. 16 are executed, and data is sequentially read from the audio data ROM 50B (steps 1601 negative, 1602 .
1603>, the stop address is reached (step 1601), a stop command is sent to the speech synthesis unit 510, and the sounding flag F1 is reset (step 1604), and the LED of the message number corresponding to is turned off (step 1605).

Next, it is confirmed that the pronunciation has ended and the message number is the first message number (step 1606, affirmative), and that the number of connections is not "1" (step 160).
7 negation) is performed, then the second message fs stored in bits 24 to 31 of the audio output command is
4o, Read start and stop addresses (
step 1609) and the second message No.
After the start address is set for the second message number (step 1610), and the LED associated with the second message number is lit (step 1611).
, the second message No. is sent to the speech synthesis section 510.

Thereafter, the process of reading audio data related to the second message number (step 1601 negative, 1602, 160
3>, Speech synthesis end processing (step 1604), LED extinguishing processing for the corresponding message number (step 1605)
After successively passing through, the message related to the end of the pronunciation is the first message.
Confirmation that this is not the th message (step 1606)
After confirming that it is the second message (step 1613, affirmative) and confirming that the number of connections is not "2" (step 1614, negative), bits 16 to 23 of the audio "°1" output command are confirmed. Start and stop reading processing for the third message stored in the bit (step 1615), start address setting processing for the third message (step 1616)
, the message No.

After the LED lighting process related to (step 1611
L A start command related to the third message is sent to the speech synthesis section 510 again.

Thereafter, the audio data reading process (step 16) is performed in the same manner.
01 negation, 1602.1603), speech synthesis end processing (
Step 1604), L related to the corresponding message No.
After sequentially going through the ED extinguishing process (step 1605), it is determined that the message related to the end of pronunciation is not the first message (step 1606 negative), and that it is not the second message (step 1613 negative), as will be described later. On the condition that the repeating process has been completed (step 1601), the three-message concatenation process is completed.

In this way, the O bit to 7 bits of the audio output command, 24
The message numbers to be sounded first, second, and third are stored in bits ~ 31 bits and 16 bits ~ 23 bits, respectively, in BCD code, and the binary numbers corresponding to the number of message concatenations are stored in 12 bits and 13 bits. Just by memorizing the code, three messages can be concatenated and pronounced at one-second intervals.

In addition, the same explanation applies even when the number of connections is "2", and
Of course, by setting the 15th priority function selection bit to "1", it is possible to interrupt messages related to no priority function as described above and to make them sound preferentially instead.

Therefore, by using this message linking function, more information can be announced by voice by arbitrarily combining the 32 messages.

In addition, as in the case of the bit-compatible command, in the case of this code-compatible command, the message NO., setting device (DIR switch 54 bits 3 to 7) provided on the audio crab unit side is used.
By using the test switch (corresponding to the second bit of the DIP switch 54) and the test switch (corresponding to the second bit of the DIP switch 54), a simulated voice output command can be given to make the specified NO message sound (step 1505 affirmative, 1512). .

Next, a case will be described in which a specified message or a plurality of messages connected to each other are repeatedly sounded a specified number of times.

In this case, in the instruction format shown in FIG. 11, a binary code corresponding to the number of repetitions is stored in 8 bits to 11 bits.

In the flowchart of FIG. 16, when each message or a series of messages connected to each other completes pronunciation (step 1607 affirmative, step 1613 negative, or step 1614
(Yes), it is determined whether the number of message pronunciations up to that point has reached the specified number of repetitions, and if it has not reached Step 1608 (No), Shun (Step 1617) has passed the pause processing for one second. >, the process of setting the start address corresponding to the #1 #1 message is performed again (step 1618).

On the other hand, if it is determined that the number of repetitions up to that point has reached the designated number of repetitions (step 160B), the sound generation process is completely terminated.

Therefore, by storing the numerical value rNJ in bits 8 to 11 of the voice output command, it is possible to repeatedly sound the same specified message N times.

As described above in detail, the operations of the system program corresponding to the code-compatible commands are as follows: (1) In response to the detection of the presence of a voice output start command, the corresponding message is immediately sounded.

(2) If message generation commands arrive one after another while any message is being generated, the messages are generated one after another at 1-second intervals in the order in which they arrive at the end of the message being generated.
3) Select the presence or absence of the priority function for each individual message or a series of linked messages using the user program,
If a message related to a priority function is generated while a message related to a non-priority function is being sounded, the message being sounded is forcibly interrupted, and a message related to a priority function that comes later is interrupted. (4) Connect and pronounce up to three specified messages in the specified order. (5) For each message or a series of connected messages, the number of repetitions is specified by a user command, and it is possible to perform an operation in which the valve is repeatedly sounded for the specified number of repetitions, and a message output test using a simulated voice output command. It is.

In the above embodiment, the original audio signal or the amplified audio signal that has been amplified to the extent that the monitor speaker can be driven is taken out from the audio output unit. Of course, the corresponding message may be pronounced from the voice output crab unit itself.

Further, in the above embodiments, the present invention is applied to a ladder diagram input type programmable controller, but it goes without saying that the present invention can be similarly applied to a flowchart input type programmable controller.

Furthermore, the above embodiment is applied to a so-called END refresh type programmable controller that detects the END command of the user program and updates the output.
It goes without saying that the present invention can also be applied to a so-called refresh-every-time programmable controller in which data corresponding to the execution result is directly transferred to the I10 unit each time a 0LIT or MOV instruction is executed.

Further, in the above embodiment, the audio output unit 5 is attached to the system bus SB directly connected to the CPU unit, but it can of course be attached to a remote system bus installed via a wired or wireless transmission line.

The above is the configuration 0 of the voice output crab knit related to this embodiment.
Having explained the functions and effects, next some examples of user programs for actually using this voice output crab knit will be explained with reference to FIGS. 17 to 23.

In the following description, it is assumed that the audio output unit is allocated to two channels, consisting of channel O and channel 1, in the input/output address space.

In the example shown in Fig. 17(a), the message ``No. 7 tank is full'' corresponding to message No. 0.10 and the message ``No. 7 tank drained completed'' corresponding to message No. 0.11 are input at 1000 and 1001, respectively. A ladder diagram is shown in which the sound is generated at the timing when the button is turned on.

In this way, it is sufficient to simply activate the output relay 10.11 corresponding to the message No. with the corresponding input 1000.degree. 1001.

In this case, as shown in FIG. 17(b), if the inputs 1000 and 1001 are turned on at different timings, the audio crab unit can recognize the ON timing of each output 10.11. As shown in FIG. 17(C), when the outputs are turned on at the same timing, priority is given to output 10 with the smaller message number.

FIG. 18(a) shows a ladder diagram in which the message "Please press the reset button" corresponding to message No. 0.10 is sounded even if either of the inputs 1000 and 1001 for failure detection is ONL.

In this case, as shown in FIG. 18(b), inputs 1000 and 1
There is no problem if the ON period with 001 does not overlap, but as shown in Fig. 18 (C), if both 1000 and 1001 partially overlap, the rise of output 10 will occur only once,
Therefore, if a failure occurs twice, the reset button must be pressed twice, which causes some inconvenience.

FIG. 19(a) shows a ladder diagram having almost the same function, using a DIFU instruction that outputs a pulse with a width of one instruction execution cycle, and improving the above-mentioned drawbacks.

In this case, as shown in FIG. 19(b), even if the ON periods of inputs 1000 and 1001 partially overlap, the DIFU 32
Since 00 and DIFtJ3201 become minute width pulses at different timings, it is possible to obtain two ON pulses as output 10, and for two failures corresponding to inputs 1000 and 1001, the message "Reset" is issued. "Please press the button" can be pronounced twice without fail.

FIG. 20(a) shows a ladder diagram in the case where when the input 1000 corresponding to the train departure command is turned ON, the message "The train is about to depart J" is sounded five times at 1-second intervals, which corresponds to the message No. 0.10.

In this case, as shown in FIG. 20 (b), in response to the ON timing of the input 1000, the output 3200 is output
Self-hold at 0, then 0.1 second clock input 630
0 turns output 10 ON 5 times, and turns ON the 5th time by C.
Make NTO count and output 3 as the count up
200 is released from self-holding.

Note that in this ladder diagram, since a 0.1 second clock is used, it is assumed that there is a slight possibility that another message will be inserted during the five counts.

Figure 21(a) is an improved version of the defect in Figure 20.
The difference from FIG. 20(a) is that the output 3202 is chattering, a high-speed pulse is obtained, and the output 10 is turned on five times at a higher speed, while the DIFU instruction is used instead of holding the input 1000 as it is. This is converted into a -H minute width pulse, and the output 3201 is thereby self-maintained.

As shown in FIG. 21(b), according to this ladder diagram, 5
Even if h1000 enters [)N again during the count,
This can be addressed immediately.

Next, the case of code-compatible instructions will be explained.

FIG. 22(a) shows messages "Valve", "1", and "Open" corresponding to message No. 01 and 02.03, respectively, in response to input 1000 being turned on.
are concatenated in order and pronounced, and the message number is generated in response to the ON of input 1001.

10.11. '! A ladder diagram is shown in which the messages ``Compressor'', ``3 units'', and ``Operate'' corresponding to 2 are sequentially connected at 1 second intervals and are sounded twice.

In this way, in the case of a code-compatible command, the data corresponding to the audio output command can be transferred to the O channel and the 1 channel using the MOV command.

What is particularly important here is that in response to the ON of inputs 1000 and 1001, which correspond to the activation conditions of the MOV command,
The first thing to do is to program the output 14 to always be activated. As mentioned above, in the case of the MOV instruction, even if the input corresponding to the activation condition turns OFF, the data at the transfer destination does not change, so if the activation condition of the MOV instruction related to the same data is repeatedly ONL. If so, its ON
This is because the timing is detected on the voice output command side via the output 14, as shown in FIG. 22(b).

FIG. 23 (a) is a countermeasure for when the ON periods of inputs 1000 and 1001 partially overlap. In this case, the DIFU instruction is used in the same manner as described above,
It is sufficient to convert the ON timing of 001 into a minute width pulse.

[Brief explanation of drawings]

FIG. 1 is a perspective view showing the overall appearance of a programmable controller to which a sound output crab unit according to the present invention is applied;
Fig. 2 is a block diagram showing the electrical configuration of the CPU unit, Fig. 3 is a general flowchart showing the system program of the CPU unit, and Figs. 4 (a) to (d) are:
Figures 5(a) to 5(d) are diagrams illustrating processing operations corresponding to the decoding results of the OUT command, Figures 5(a) to 5(d) are diagrams explaining the processing operations corresponding to the decoding results of the MOV command, and Figure 6 is a diagram illustrating the processing operations corresponding to the decoding results of the MOV command. FIG. 7 is a block diagram showing the electrical configuration of the sound output unit, and FIG. 8 is a perspective view showing the appearance of the output unit. 9 is a memory map showing the contents of the audio data ROM. FIG. 10 is a diagram showing the instruction format of bit-compatible instructions. FIG. 11 is a diagram showing the instruction format of code-compatible instructions. 16 to 16 are flowcharts showing the configuration of a system program for controlling the voice output crab unit, and FIGS. 17 to 21 are diagrams showing examples of user programs for operating the voice output crab unit with bit-compatible instructions. FIGS. 22 to 24 are diagrams showing examples of user programs for operating the voice output crab knit with code-compatible commands. 1...Programmable controller 3...CPL
I units 1 to 5...Audio output unit 52...Message number, display LED 53...Audio signal extraction jack 54...DIP switch 57...Audio memory card 501...MPLI 502... - System program ROM 503... Working RAM 504, 505, 506... Latch unit 508... Audio data ROM 509... Jumper wire 510... Audio synthesis unit 511... Low pass filter 512... Transformer 513... Monitor amplifier 514... Selector switch DB... Data bus AB... Address bus CB... Control bus

Claims (1)

[Claims] (1) A data importing means for importing data sent to the unit onto the system bus of the programmable controller; Deciphering the imported data as an audio output command, and connecting each unit to be output. a command decoding means for detecting voice content, concatenation order, and presence or absence of an output start command; in response to the presence of the output start command being detected, a pronunciation command related to each detected voice content is transmitted to the detected concatenation; A sound output unit for a programmable controller, comprising: sound generation command means that sequentially issues sound signals each time the sound production ends; sound signal generation means that generates a corresponding sound signal in response to the sound sound command; .