JPH1079824A - Voice recording and reproducing method for facsimile equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide voice recording and reproducing method of facsimile equipment which is unaffected by the capacity of a memory for sound recording and holds messages even when powered off. SOLUTION: Facsimile equipment which has an automatic answering function is so constituted as to print 22 voice data obtained by digitally converting 24 and compressing a voice signal received through a communication line on a recording form, in the form of a bar code shape corresponding to its 0's and 1's, reading the voice data printed on the recording form out and storing them in the memory 12, and restore and reproduce the voice signal from the read voice data. Those processes are controlled by a CPU 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a facsimile apparatus, and more particularly to a method for processing voice data by a facsimile apparatus having an automatic response function.

[0002]

2. Description of the Related Art Generally, a facsimile apparatus is an OA apparatus for exchanging various documents, photographs, and the like with a remote party over a telephone line. With the development of industry and diversification of information, OA equipment has been increasingly required as a composite office equipment having not only one simple function but also various functions. Such a demand for a complex function has led to the development of a facsimile machine as a complex office equipment including an automatic answering function and a printing function of a personal computer.

The automatic response function includes an analog automatic response method using a magnetic tape as a recording medium of an audio signal, and a DRA.
M (Dynamic Random Access Memory) and ARAM (Audio R
digital automatic response system using a semiconductor memory such as andom Access Memory). recently,
With the rapid development of semiconductor memory, an automatic transponder that can easily write and read data has been developed. However, in an automatic transponder using such a semiconductor memory, the power is turned off in consideration of the memory price and the like. Since volatile memory elements that lose data are predominant, there is a problem of temporal storage of data and a limitation of storage capacity for storing a large amount of data.

[0004]

The present invention, which focuses on the prior art as described above, enables a facsimile apparatus to use a recording sheet as a storage medium for voice, and has the above-mentioned temporal storage characteristics and storage capacity. It tries to make up for the problem.

[0005]

SUMMARY OF THE INVENTION For this purpose, according to the present invention, voice data obtained by digitally converting and compressing a voice signal received through a communication line is printed on a recording sheet in a predetermined form. A voice recording method using a facsimile apparatus with an automatic response function, characterized by reading voice data printed on recording paper in such a predetermined recording form, storing it in a memory, and restoring a voice signal from the read voice data. Provided is a voice reproducing method for a facsimile apparatus with an automatic response function, characterized by reproducing.

In the above-described audio recording method, it is preferable that the audio data is printed in the form of a bar code, that is, a space and a bar corresponding to the logic of the data. Also, the recording paper is provided with an upper margin, a lower margin, a left margin, and a right margin so that the audio data is printed. Then, before the audio data recording unit that prints the audio data, an information recording unit that displays information about the audio data to be recorded may be printed. The information recording unit prints a skew reference line, an information recording area, and an audio data recording start line. The information recording area contains the title of the print content, the date and time when the audio data was received,
It is good to print the number of pages. The audio data recording unit prints the data lines of the space and the bar and the margin between the data lines. The data line has a start bit, a data bit,
It is preferable to print in the order of the end bits. The start bit is 1-bit data printed on the bar, and the data bits are such that 0 is a space and 1 is a bar. The recording unit size of these start bit, data bit and end bit is 8 horizontally.
It is 4 pels for a recording resolution of pels / mm, and 5 lines vertically for a recording resolution of 7.7 lines / mm.
The margin between lines in this case is 7.7 lines / mm vertically
It is better to set 3 lines for the recording resolution.

In the above audio reproducing method, it is preferable to compensate for a vertical skew caused by feeding of the recording paper when reading the printed audio data.

Further, according to the present invention, as a method of reproducing sound by a facsimile apparatus having an automatic response function, a step of scanning a recording sheet on which sound data is printed in a predetermined recording form, and a step of scanning image data obtained by the scanning. A step of compensating for the vertical skew, a step of converting the image data with the compensated vertical skew into audio data, a step of synthesizing the converted audio data as an audio signal and reproducing the audio signal.
Is provided. The vertical skew compensation step includes a step of detecting a skew reference line from image data obtained by scanning, a step of detecting a vertical skew compensation value based on the detected skew reference value, and a step of detecting the vertical skew compensation value. And compensating the vertical skew of the image data obtained by scanning if it is smaller than the predetermined reference value. Further, in this case, if the detected vertical skew compensation value is larger than a predetermined reference value, a process of re-scanning the recording paper may be further performed. Further, it is preferable that a window is placed on the image data in which the vertical skew has been compensated, and a step of performing secondary vertical skew compensation for each of five lines constituting one data line and five lines above and below the line is further performed. The process of converting the audio data into audio data includes a process of detecting a start bit and a data bit from the image data subjected to the vertical skew compensation, and a process of detecting a vertical position of a data line after detecting the data bit and converting the data line into audio data. And to carry out. Further, a step of counting the number of continuous horizontal pixels by detecting data bits and compensating for a horizontal error, and a step of counting the number of bits of the detected data and detecting an end bit after compensating for the horizontal error are further performed. Good to do. When the end bit is detected, it is preferable to check the presence or absence of a parity error, and to further perform a process of re-scanning the recording sheet when the parity error is detected. The horizontal error compensation process includes a process of dividing the count value of the number of continuous horizontal pixels by 4 to detect a remainder, and a process of converting audio data according to a quotient obtained by dividing the count value by 4 when the detected remainder is 0. The process to be performed and if the detected remainder is 1
When 1 is subtracted from the count value and divided by 4, the voice data is converted in accordance with the quotient, and when the detected remainder is 3, 1 is added to the count value.
And converting the voice data according to a quotient obtained by adding 2 to the count value and dividing by 4 if the previous pixel compensation value is smaller than 0, when the detected remainder is 2, Performing the conversion and, if the previous pixel compensation value is greater than 0, subtracting 2 from the count value and dividing by 4 to perform audio data conversion.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram of a facsimile apparatus having an automatic response function. The CPU 10 controls the general operation of the facsimile according to the set program,
In particular, control for recording received audio signals on recording paper,
Then, control for scanning the recording paper and reproducing the audio signal is performed. The memory 12 stores audio data and a program for recording and reproducing the audio data, and temporarily stores data generated when the program is executed. The OPE (OPERATING PANNEL) 14 has a large number of keys, sends key data by key operations to the CPU 10, and displays an operation state according to the control of the CPU 10. The scanner 16 scans the document, binarizes the document image, and inputs it to the CPU 10. Modem 18
Under the control of the PU 10, the transmission data is analog-modulated and output, and the analog reception data is demodulated and output. NC
The U (Network Control Unit) 20 operates under the control of the CPU 10 and forms a telephone line communication loop to take charge of an interface between the modem 18 and the telephone line.

The printer 22 prints data on recording paper under the control of the CPU 10. Codec
-decoder) 24 converts an analog audio signal input through the NCU 20 into a digital signal, and converts the analog audio signal into a digital signal.
And converts the digital audio signal into an analog audio signal and outputs it to the NCU 20. The audio amplifier 28
An analog audio signal is received from the codec 24, amplified to a predetermined level, and output to the speaker 30. Voice processing unit 2
Reference numeral 6 denotes a DSP (Digital Signal Processor) which can be realized by using, for example, D63754A of DSP Co., and synthesizes audio data stored in the memory 12 under the control of the CPU 10 and outputs the synthesized data to the codec 24. Also, the digital audio signal from the codec 24 is compressed and provided to the memory 12.

FIG. 2 is an explanatory view showing an image reproducing area in the facsimile apparatus of this embodiment, FIG. 3 is an explanatory view showing a recording area for A4-size recording paper in the facsimile apparatus of this embodiment, and FIG. FIG. 4 is an explanatory diagram showing an example of the embodiment. FIG. 5 is a print format diagram of audio data, where a is a data line, b is a margin between lines, c is a left margin, d is a start bit, e is a data bit, f is an end bit (parity bit), and g is The right margin. 6 is a detailed recording format diagram of audio data, and FIG. 7 is an explanatory diagram showing one form of audio data printed on recording paper. FIG. 8 is a control flowchart for recording audio data on recording paper,
FIG. 9 is a control flowchart for reproducing audio data printed on recording paper. 10 and 11
Is a flowchart of a vertical image skew compensation subroutine, FIG. 12 is a flowchart of an audio data restoration subroutine, FIG. 13 is a flowchart of a horizontal error compensation subroutine, and FIG. 14 is a flowchart of a double horizontal error compensation subroutine.

The control process of the preferred embodiment of the present invention will be described in detail with reference to FIGS.

ISO (International Standards Origani)
zation) In a Group 3 facsimile machine that can transmit and receive A4 size (210 mm x 297 mm) originals,
CITT (International Telegraph and Telephone Advisory Committee) Recommendation
The image reproduction area recommended in T.4 takes into consideration horizontal loss and vertical loss that can occur during printing and scanning, and as shown in FIG.
It is recommended to guarantee a range of 1.46 mm. Tables 1 and 2 below show CCITT REC. T.
4 shows the horizontal loss and the vertical loss that can occur when reading and recording an A4 size document of a Group 3 facsimile apparatus recommended in Table 4, respectively. Table 3 below shows the resolution of recording and reading for horizontal and vertical.

[Table 1]

[Table 2]

[Table 3]

In reading an image, light from a light source is reflected by a document, and the reflected light is detected and read by a reading sensor.
As the reading sensor, a CCD (Charge Coupled Device) chip, a CIS (Contact Image Sensor), or the like is mainly used, and is composed of optical sensors arranged at intervals such as the horizontal resolution shown in Table 3. ISO A4 size (210mm x 2
An optical sensor for reading an original of 97 mm) generally has a reading width of 216 mm so as to be able to read even a letter size (216 × 297 mm) original, and the reading sensor has an interval of 8 / mm for 216 mm in the horizontal direction. Have a total of 1728 optical sensor arrays. Note that a fluorescent lamp or an LED array is used as a light source. The printing method is classified into a thermal recording system and a general paper system according to the recording paper to be printed. The vertical resolution for recording is 3.85 lines / mm as a standard, and 7.7 lines as an option.
/ mm and a resolution of 15.4 lines / mm, and most Group 3 facsimile machines have 7.7 l
It helps to record up to ines / mm vertical resolution.

The facsimile apparatus of this embodiment having such a resolution is provided with an automatic answering function, and the automatic answering function is the same as the technology normally used in an automatic answering telephone. The facsimile apparatus of the present embodiment stores a voice signal received from a caller by such an automatic answering function in a voice memory, and converts the data stored in the voice memory into a code form as shown in FIG. It is possible to print on recording paper. A process of temporarily storing the audio signal in the audio memory, converting the audio signal into a code form as shown in FIG. 7, and printing it on recording paper will be described with reference to FIG.

First, at step 101, the CPU 10
20 to check whether a ring signal (call) is detected (R
ING DETECTED), the process proceeds to step 102 when a ring signal is detected. At step 102, the CPU 10 sets the voice receiving mode (VOICE RECEIVING MODE) according to the key input section of the OPE 14.
It is checked whether or not is set, and if the voice receiving mode is not set, the process proceeds to step 103. In step 103, the function according to the set mode is performed (PERFORM FUNCT
ION AS CORRESPONDING MODE). That is, if the TEL mode is set, the call is made with the handset hooked off and the FA is set.
When the X mode is set, the fax document is received and the AN
Fax is sent automatically when S / FAX mode is set.
Receive the document. On the other hand, if the voice receiving mode is set in step 102, the process proceeds to step 104, and the CPU 10
Controls the voice processing unit 26 to read and synthesize voice data stored in the memory 12, and sends out a voice guidance message (OUT GOING MESSAGE: OGM). The transmitted voice guidance message is converted into an analog signal via the codec 24 and transmitted to the calling party via the NCU 20.

If there is a voice input from the caller after transmitting the voice guidance message, the voice signal is received (RECIEVE VOICE DATA) in step 105, and the received voice signal is NCU2.
0 is sent to the codec 24, converted into a digital signal, and input to the audio processing unit 26. Then, at step 106, the CPU 10 controls the audio processing unit 26 to compress the digitally converted audio data and store it in the memory 12 (COM
PRESS AND STORE RECEIVED VOICE DATA). Then 107
Check that all audio data reception has been completed at the stage (R
If not completed, the process returns to step 105 and repeats the same steps to receive and store voice data. If completed, the process proceeds to step 108, where the CPU 10 reads the audio data stored in the memory 12 and controls the printer 22 to perform printing (READ STOR).
ED VOICE DATA AND PRINT READ VOICE DATA).

FIG. 3 shows a print format for printing audio data on A4 size recording paper. FIG.
In the above, a is an upper margin, b is a lower margin, c is a left margin, d is a right margin, and e is an information recording unit.
f is a recording unit of audio data. That is, the print output part includes an information recording unit e for recording information such as a title and a date of the audio data to be recorded, and an audio data recording unit f for recording the audio data. The margins a to d are determined in consideration of an effective printing area and an effective scanning area given by the structural characteristics of the facsimile machine.

As shown in FIG. 4, the information recording section includes a skew reference line a, a title of print contents, a reception date and time of audio data, an information recording area b for storing the number of pages of recorded data, and an audio data recording start. It consists of line c.

The audio data recording section has the format shown in FIG. 5 and has a space (white) for binary data and a bar (black) for 1 data for binary audio data.
Is applied and recorded. And 0 and 1
One line (one line) is recorded in the space and bar notation corresponding to the combination of audio data, and if the recording of continuous audio data exceeds one line, a new line is started and the audio data is continued to the next line. To record. When recording audio data on the next line, a margin b between lines is inserted between each line in order to distinguish the audio data from the previous line. One data line a obtained in this way has a left end margin c and a right end margin g as margins, and a space between the left end margin c and the right end margin is defined as a start data d, a data bit e, and an end bit f. Is done. The start bit d is displayed as a 1-bit bar and represents the beginning of one line of audio data. The end bit f is a bit indicating the end of one line of audio data and is a parity bit. Even parity (EVEN PARITY) per data line a is detected so that a bit error that may occur when reading audio data can be detected and corrected. ).

As shown in FIG. 5, one of the audio data to be printed
The size of the bit recording unit is applied in consideration of the recording resolution of the facsimile apparatus, the image reading resolution, and the skew of the original at the time of reading the image in the horizontal and vertical directions. In the present embodiment, the size of a 1-bit recording unit for the horizontal and vertical resolutions of the printed audio data is as shown in FIG. That is, the size of the recording unit for one bit is 8 pels / mm in the horizontal (Horizontal) resolution (rezolut
For example, 4 pels is applied as a 1-bit recording unit for ion, and 5 lines is applied as a 1-bit recording unit for a resolution of 7.7 lines / mm in Vertical.
Also, the margin between lines b (space between data lines)
For, three vertical lines were applied. FIG. 7 shows an actual overall image of the sound data stored in the memory 12 printed on the recording paper by the printer 22 in this manner.

Next, a process in which the audio data is printed out on a recording sheet in the form of a space bar as described above, and the original is scanned to reproduce the audio will be described with reference to FIGS. .

At step 201 in FIG. 9, the CPU 10
Voice playback key (VOICE RESTORING)
KEY) is entered. When the sound reproduction key is input, the process proceeds to step 202 and the CPU 10
Checks whether a document is inserted in the platen (DOCUM
ENT INSERTED). If there is no original, the process proceeds to step 203 to generate an alarm sound (GENERATION OF ALARM TONE), and then proceeds to step 204. At step 204, the CPU 10 displays a document loading request message for requesting document loading on the display unit of the OPE 14 (DISPLAY DOCUMENT LOADIN).
G REQUEST MESSAGE).

If the insertion of the document is confirmed at step 202, the process proceeds to step 20.
Proceeding to five steps, the CPU 10 controls the scanner 16 to read the original image and obtain binarized image data.
The reading of the document image at this time uses a method in which the document transport roller is driven by a stepping motor to transfer the document on a fixed reading device and to take in image data from the document image. As described above, in the mechanism for transporting the original on the reading device by the original transport roller,
Since a skew phenomenon of the document occurs according to the mechanical characteristics, in some cases, it may not be possible to reproduce sound from the read binary image data. Therefore, 206
At this stage, the CPU 10 performs vertical skew compensation of the obtained image data (CORRECTION OF VERTICAL SKEW).

In order to compensate for the vertical skew of the image data, there must be a reference for sensing the skew information of the image data. In this embodiment, however, the upper end of the original on which the audio data is recorded is shown in FIG. Skew is compensated using the reference line a of the information recording unit provided as described above. The process of this vertical skew compensation is shown in FIGS.

First, at step 301, the CPU 10 stores the first scanned image data in the memory 12 (S
TORE IMAGE DATA IN MEMORY). In this example, at this time, image data for 28 lines of the information recording unit is stored in the image storage area of the memory 12. Then 30
In two steps, the CPU 10 detects a compensation reference line of the stored image data (DETECT CORRECTION REFERENCE LINE), and then in step 303, detects vertical skew information for each horizontal pixel, that is, a compensation reference value for each horizontal pixel (DETECT CORRECTION VALUE).
FOR EACH PIXEL) and proceed to step 304. Memory 1
2 is the maximum compensable reference value for the vertical skew of the image data stored in CCITT REC. T. 4 to compensate for the vertical skew of up to 3.6 mm. The vertical skew is detected by using the vertical position value of the compensation reference line as a skew value and detecting a compensation value. 30
In four steps, the CPU 10 checks whether the detected vertical skew compensation value exceeds the maximum compensable reference value (SKE
If W ≧ REFERENCE VALUE), and exceeds the maximum compensable reference value, the process proceeds to step 305. At step 305, the CPU 10 sounds a warning sound or displays a message on the display unit of the OPE 14 to notify the user, prompts the user to reinsert the document, and starts reading the document image again (REQUEST OF RESCAN IMAGE).

On the other hand, if the skew compensation value does not exceed the maximum compensable reference value, the CPU 10 compensates the skew for the 28 lines of image data stored in the memory 12 (CORRECT SKEW OF STORED IMAGE DATA) at step 306. )
. When the vertical skew is compensated for in this manner, the data skew appearing in the image data to be further transported by the original transport roller and read from the original is skewed continuously during the entire process of the original passing through the reading section. Since the skew appears cumulatively, the following method is used to accurately restore all data by skew compensation using the compensation reference line located at the upper end of the document.

That is, after confirming the start line c shown in FIG. 4, in step 307, the CPU 10 compensates for the vertical skew of the image data of the first line and stores it in the image window buffer for conversion into audio data. (STORE CORRE
CTED 1'ST LINE DATA IN BUFFER). Next, at step 308, it is checked whether image data for 15 lines has been stored in the buffer for the image window (BUFFER DATA = 15 LINES).
If 5 lines of image data are not stored, 312
Proceed to stage. In this step 312, the image data stored in the image window buffer is the last line (LAST LINE).
Is checked, and if it is the last line, the vertical skew compensation ends. If it is not the last line, the process proceeds to step 313, and the CPU 10 reads the image data of the next one line (IMAGE DATA SCANNING OF 1 NEW LINE).
Proceed to step 4. At step 314, the CPU 10 compensates for the vertical skew of the read image data (SKEW CORRECTION
(SCANNED IMAGE DATA) 315 is stored in the memory 12 (STORE IMAGE DATA IN MEMORY). Thereafter, at step 316, the CPU 10 stores the image data in which the vertical skew has been compensated in the image window buffer for conversion into audio data (STORE CORRECTED 1 LINE DATA IN BUFFER).
Returning to step 08, the above process is repeated to compensate for the vertical skew of the read image data.

The area for storing 15 lines of the image window buffer is composed of 5 lines.
For one data line (FIG. 5A and FIG. 6), the value is set so that the vertical skew can be compensated for each of the five lines vertically. While compensating for the vertical skew of the image data thus read, the process proceeds to step 207 (step 309) in FIG. 9 and the CPU 10 restores the audio data.

The restoration process of the audio data will be described in detail with reference to FIG. At the stage 401, each pixel (HO = 1-782) from the first pixel to the 1728th pixel on the horizontal side (X = 1-1782)
R.BIT) is compared with 15 line data (BUFFER LINE) stored in the buffer (Y = 1-15). And that one
As a result of the comparison of the five line data, if black data is continuously detected in five lines, it is determined that a bit has been detected, and the process proceeds to step 402, where four or more detected black data are continuously detected horizontally. If it is located, it is determined that the start bit has been detected (DETECT START BIT). Then at 403 stage CP
U10 detects the data bit by checking the number of black and white pixels continuously located in the horizontal direction after the start bit is detected (DETECT DATA BIT).

In accordance with the detection of the data bit, the CP
U10 analyzes the position of the line where data exists from 1728 horizontal pixels for 15 lines of the image window buffer (DETECT LINE POSITION OF).
DATA IN BUFFER). When the position of the horizontal pixel line is detected by this analysis, the values of the uppermost vertical line (Ymin) and the lowermost vertical line (Ymax) where the data is located are analyzed. And the top vertical line as in Equation 1 below
The center value (Ycen) of the data position is calculated by averaging the (Ymin) value and the lowest vertical line (Ymax) value. Next, at step 411, the CPU 10 calculates the center value (Ycen) as in Equation 2.
To the buffer line shift value (Yshift) (DETEC
T SHIFT VALUE OF BUFFER LINE).

## EQU1 ## Ycen = (Ymin + Ymax) / 2

## EQU2 ## Yshift = Ycen

The shift of the buffer line shifts a window for reading data, and shifts a buffer area by a shift value calculated from top to bottom with respect to a line of image data. In other words, 15
If the line is accumulated, the process proceeds to step 309 to perform a voice data conversion process (VOICE DATA CONVERSION). If the line shift value of the buffer is determined in step 310 as described above (LINE SHIFT VALUE OF BUFFER), step 311 Executes buffer shift (BUFFER SHIFT). When the buffer moves, new image data is continuously read and stored in the buffer to read the audio data. And 40
When a data bit is detected in three steps, the process proceeds to step 404 to detect the vertical position value of the data, and then proceeds to step 411.

On the other hand, if a data bit is detected in step 403, step 405 is executed. Since one bit of data is composed of four horizontal pixels as shown in FIG. 6, the number of continuous pixels is counted. (COUNT CONTINUOS PIXEL), 4
At step 06, the CPU 10 compensates for the horizontal reading error (COR
RECT HORIZONTAL ERROR). The process of compensating for this horizontal reading error will be described with reference to FIG.

First, at step 501, the CPU 10 divides a value obtained by counting the number of continuous black (1) or white (0) pixels by 4 (COUNT ÷ 4), and at step 502, the remainder (REMAIND).
Check ER). At this time, the remainder is 0 (REMA
If (INDER = 0), the process proceeds to step 510, where the audio data conversion of 0 or 1 is performed by the quotient (QUOTIENT) obtained by dividing the value obtained by counting the number of continuous pixels by 4. If the remainder is 1 (REMAINDER = 1) at step 504, the process proceeds to step 505 to subtract 1 (COUNT = COUNT-1) from the count value of the number of continuous pixels.
Proceeding to step 510, audio data conversion of 0 or 1 is performed by a quotient obtained by dividing the subtracted value by 4. Further, if the remainder is 3 (REMAINDER = 3) in step 508, the process proceeds to step 509, where 1 is added to the count value of the number of continuous pixels (COUNT = COUNT + 1). The audio data conversion of 0 or 1 is performed only by the quotient obtained by dividing by 4. On the other hand, if the remainder is 2 (REMAINDER = 2) in step 506, the process proceeds to step 507 to perform the double compensation subroutine, and then proceeds to step 510. FIG. 1 shows the process of this double compensation subroutine.
This will be described with reference to FIG.

In this process, assuming that the current continuous pixel is the Nth continuous pixel, the compensation value is determined with reference to the (N-1) th and (N + 1) th continuous pixel numbers. First, in step 601, the current continuous pixel is set as the Nth continuous pixel (CURRENT CON
TINUOUS PIXEL = N'TH CONTINUOUS PIXEL), and 60
Check the (N-1) th continuous pixel compensation value in two steps (N
-1'TH CONTINUOUS PIXEL CORRECTION VALUE). At this time, if the compensation value is 0 at step 603, (CORRECTION VALUE
= 0), and proceeds to step 604 to check the (N + 1) th continuous pixel compensation value. As a result, if the (N + 1) th continuous pixel compensation value is also 0, retry is performed at step 305 in FIG. In addition, as a result of checking the N-1th continuous pixel compensation value in step 602, if the continuous pixel compensation value is smaller than 0 in step 605 (CORRECTION VALUE <0), the process proceeds to step 606. This corresponds to the case of the 505-step subtraction. 60
In the sixth stage, 2 is added to the current count value of the number of continuous pixels (COUNT = COUNT + 2). Thereafter, in step 510, audio data conversion of 0 or 1 is performed only for the quotient obtained by dividing the count value obtained by adding 2 by 4. Further, as a result of checking the N-1th continuous pixel compensation value in step 602, if the continuous pixel compensation value is greater than 0 in step 607 (CORRECTION VALUE> 0), 6
Proceed to step 08. This corresponds to the case of addition in 509 steps. In step 608, 2 is subtracted from the current count value of the number of continuous pixels (COUNT = COUNT-2). Thereafter, at step 510, the count value obtained by subtracting 2 is divided by 4, and the quotient is used to perform audio data conversion of 0 or 1.

When data bits are detected in step 510 in this way, the number of bits detected in step 511 is checked to determine whether or not the number of bits recorded in one data line has been detected. That is, it is searched whether M is 0 (M = 0). If M is not 0, 1 bit of audio data of 0 or 1 obtained by the conversion is stored in the memory 12.
(STORE ONE BIT VOICE DATA) and proceed to step 513. 5
At step 13, the CPU 10 decreases M by 1 (M = M−1), 51
Returning to one stage, the above process is repeated until the number of bits recorded in one data line is completely detected.

After compensating the horizontal error, the CPU 10 counts the number of bits of the audio data detected from the data line at step 407 in FIG. 12 (COUNT BIT NUMBER OF VOI).
In step 408, it is checked whether the end bit (END BIT) has been reached from the result of counting the number of bits. If the end bit has not been reached, the process returns to step 403, and the process up to step 407 is repeated until the end bit is reached. That is, the end bit is a bit indicating the completion of detection of all bits recorded in one data line. When the end bit is detected in step 408, the process proceeds to step 409, and the CPU 10 checks the number of 1 (black) bits detected from the current data line to search for occurrence of a parity error (PARITY ERROR). If a parity error has occurred, the CP
U10 prompts rereading of the original by sounding an alarm sound or displaying a message on the display unit of OPE 14 (REQUEST OF RESCAN IMAGE). On the other hand, if there is no parity error, the average of the uppermost vertical line (Ymin) value and the lowermost vertical line (Ymax) value is calculated, and the center value (Y
After calculating cen), the center value (Ycen) is set to the shift value (Yshift) of the buffer line in step 411.

When the image data is read for each data line and converted into audio data as described above, 2 in FIG.
At step 08, the obtained audio data is stored in the memory 12.
(STORING OF CONVERTED DATA), proceed to step 209. 2
At step 09, the CPU 10 reads the audio data from the memory 12, synthesizes the audio by the audio processing unit 26, converts the audio data into an analog signal through the codec 24, and the analog-converted audio signal is amplified by the audio amplifier 28 to the speaker 30. Sent and played back (SYNTHESIZE CONV
ERTED DATA INTO VOICE AND RESTORE VOICE INTO VOICE
SIGNAL).

The facsimile apparatus of the present embodiment was connected to a personal computer, and an experiment of voice recording / reproduction was performed. The experimental recording data uses random data using a random function, and the evaluation of the experiment evaluates the recording capacity of the data on one recording sheet of ISO A4 size for the data recording algorithm and the reproduction rate for the reproduction algorithm. I went by that. The read / reproduce rate was evaluated by comparing the bit error between the restored data and the initial recorded data.

The data recording capacity for one ISO A4 size recording sheet is determined by the recording area for recording the data and the size of the recording unit. The recording area for the ISO A4 size recording paper is divided as shown in FIG. 3, and the size of each area applied in this experiment is as shown in Table 4 below.

[Table 4]

The horizontal and vertical sizes of the data recording unit bits and the size of the interline space (margin) are as shown in Table 5 below.

[Table 5]

Accordingly, the data recording capacity of the ISO A4 size recording paper is as shown in Table 6 below.

[Table 6]

As shown in Table 6, 392 bits of data per data line can be recorded on A4 size recording paper, and 269 data lines can be recorded on one recording paper. Approximately 13.1 Kbytes of data can be recorded on one A4 size recording sheet. When storing the voice signal in the memory 12, the voice processing unit 26 for performing the digital automatic response converts the voice data modulated by the PCM method into the 8-bit digital signal at the sampling of 8 KHz using the 4 Mbit memory for about 15 to 15 bits. It provides a compression rate that can store 17 minutes worth of audio data.
Therefore, the memory 1 is implemented by using the D6375A DSP chip.
The audio data stored in No. 2 can record a capacity of about 25 seconds per one ISO A4 size recording sheet.

FIG. 7 shows an example of print data in which data actually stored on a recording sheet of data stored in a memory is enlarged and displayed at 208%. In the experiment of the reproduction algorithm, 100 kinds of random data were recorded on the recording paper,
The recorded data was read ten times each to reproduce the data. In addition, plain paper (Plain
Paper) and thermal recording paper were used, and the reading characteristics of each recording paper were compared. In the evaluation of the experiment, the number of occurrences of parity errors that occurred during data reproduction and the read / reproduction rate by comparing bit errors between reproduced data and initial recording data were evaluated. Then, as shown in Table 7 below, 1000 reading / reproducing experiments were performed for each recording paper, with reading / reproducing 10 times for 100 kinds of recording data, respectively.

[Table 7]

As a result of such an experiment, as shown in Table 7, six reading errors occurred on ordinary paper and 0.6
% Of read / write error rate, and 14
A read error occurred 1.4 times, indicating a read error occurrence rate of 1.4%. However, when a re-test was performed so that the sheet was correctly read when a reading error occurred, no reading error occurred. The reason for the difference in the read / reproduce rate between recording papers is that the skew of the image data during reading is small due to the characteristics of the recording paper in general paper, and the image skew is large during reading due to the roll phenomenon of the recording paper in thermal recording paper. is there. As described above, the algorithm of the present invention for data recording and reproduction has a difference in the reproduction result due to the recording paper and the mechanical characteristics of the system, but this is not practically problematic.

[0047]

According to the present invention, in a facsimile apparatus having an automatic answering function, a received voice is printed in a predetermined data form, stored on a recording paper, and the recording paper is scanned to reproduce the voice. Therefore, all received voices can be stored and played back regardless of the capacity of the voice memory. Therefore, there is an advantage that a large amount of audio data can be stored beyond the limit of the memory capacity, and that the data can be stored and reproduced semi-permanently irrespective of the power on / off.

[Brief description of the drawings]

FIG. 1 is a block diagram of a facsimile apparatus having an automatic response function.

FIG. 2 is an explanatory diagram illustrating an image reproduction area of the facsimile apparatus.

FIG. 3 is an explanatory diagram exemplifying a recording area for A4 size recording paper of the facsimile apparatus.

FIG. 4 is an explanatory diagram showing a specific example of an information recording unit e in FIG. 3;

FIG. 5 is an explanatory diagram showing a recording format of audio data.

FIG. 6 is an explanatory view showing more details of a recording format of audio data.

FIG. 7 is an enlarged view showing an example of a recording sheet on which audio data is recorded.

FIG. 8 is a control flowchart for explaining a voice recording process.

FIG. 9 is a control flowchart for explaining a sound reproducing process.

FIG. 10 is a flowchart of a vertical image skew compensation subroutine.

FIG. 11 is a flowchart following FIG. 10;

FIG. 12 is a flowchart of an audio data restoration subroutine.

FIG. 13 is a flowchart of a horizontal error compensation subroutine.

FIG. 14 is a flowchart of a double horizontal error compensation subroutine.

Claims (22)

[Claims] 1. A voice recording method using a facsimile apparatus having an automatic response function, wherein voice data obtained by digitally converting and compressing a voice signal received through a communication line is recorded on a recording paper in a predetermined form. A voice recording method characterized by being printed. 2. The audio recording method according to claim 1, wherein the audio data is printed in a form of a space and a bar corresponding to the logic of the data. 3. The audio recording method according to claim 2, wherein the audio data is printed by providing an upper margin, a lower margin, a left margin, and a right margin on the recording paper. 4. An audio recording method according to claim 3, wherein an information recording section for displaying information on the audio data to be recorded is printed before the audio data recording section for printing the audio data. 5. The audio recording method according to claim 4, wherein the information recording unit prints a skew reference line, an information recording area, and an audio data recording start line. 6. The voice recording method according to claim 5, wherein the information recording area prints the title of the print content, the date and time of receiving the voice data, and the number of pages. 7. The audio recording method according to claim 4, wherein the audio data recording unit prints the data using space and bar data lines and an interline margin between the data lines. 8. The audio recording method according to claim 7, wherein the data line is printed in the order of a start bit, a data bit, and an end bit. 9. The start bit is a bar printed 1
9. The audio recording method according to claim 8, wherein the data is bit data. 10. The audio recording method according to claim 8, wherein the data bits are printed such that 0 is a space and 1 is a bar. 11. The recording unit size of a start bit, a data bit, and an end bit is 4 pels for a recording resolution of 8 pels / mm horizontally and 5 lines for a recording resolution of 7.7 lines / mm vertically. The voice recording method according to claim 8, wherein 12. A margin between lines is 7.7 li vertically.
8. The audio recording method according to claim 7, wherein 3 lines are set for a recording resolution of nes / mm. 13. A voice reproducing method using a facsimile apparatus having an automatic response function, wherein voice data printed on a recording sheet in a predetermined recording form is read and stored in a memory, and a voice signal is converted from the read voice data. A sound reproducing method characterized by restoring and reproducing. 14. The audio reproducing method according to claim 13, wherein a vertical skew caused by a recording paper feed is compensated when reading the printed audio data. 15. A method for reproducing sound by a facsimile apparatus having an automatic response function, comprising the steps of scanning a recording sheet on which sound data is printed in a predetermined recording form, and determining a vertical skew of image data obtained by the scanning. Audio reproduction, comprising the steps of: compensating, converting the image data compensated for the vertical skew into audio data, and reproducing the converted audio data as an audio signal by synthesizing the audio data. Method. 16. The vertical skew compensation step includes the steps of: detecting a skew reference line from image data obtained by scanning; detecting a vertical skew compensation value based on the detected skew reference value; 16. The audio reproducing method according to claim 15, further comprising: compensating for vertical skew of image data obtained by scanning if the vertical skew compensation value is smaller than a predetermined reference value. 17. The audio reproducing method according to claim 16, further comprising the step of re-scanning the recording paper if the detected vertical skew compensation value is larger than a predetermined reference value. 18. The method according to claim 16, further comprising the step of covering a window on the image data in which the vertical skew has been compensated, and performing secondary vertical skew compensation in each of five lines constituting one data line and five lines above and below the five lines. The described audio playback method. 19. The audio data conversion process comprises the steps of detecting a start bit and a data bit from image data subjected to vertical skew compensation, and detecting a vertical position of a data line after detecting the data bit. 19. The audio reproducing method according to claim 15, further comprising the steps of: 20. A step of counting the number of continuous horizontal pixels by detecting data bits and compensating for a horizontal error, and after compensating for the horizontal error, counting the number of bits of the detected data and detecting an end bit. 20. The audio reproducing method according to claim 19, further comprising: 21. The audio reproducing method according to claim 20, further comprising the step of: confirming the presence / absence of a parity error when detecting the end bit, and re-scanning the recording paper when the parity error is detected. 22. A horizontal error compensating process includes a process of dividing the count value of the number of continuous horizontal pixels by four to detect a remainder, and a quotient obtained by dividing the count value by four when the detected remainder is zero. And performing audio data conversion according to a quotient obtained by subtracting 1 from the count value and dividing by 4 when the detected remainder is 1.
When the detected remainder is 3, the count value is set to 1
And performing a voice data conversion in accordance with a quotient obtained by dividing by 4, and when the detected remainder is 2, if the previous pixel compensation value is smaller than 0, add 2 to the count value to add 4 to the count value.
And performing a voice data conversion according to a quotient obtained by subtracting 2 from the count value if the previous pixel compensation value is greater than 0 and dividing by 4 if the previous pixel compensation value is greater than 0. The described audio playback method.