JP2000331112A - Accounting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an accounting system provided with an input system which can minimize the use of a keyboard and can greatly lighten the burden of inputting work to an operator. SOLUTION: An accounting system optically reads a slip 12, displays the read information on a screen 41, and, after collating the displayed contents of described items, stores a prescribed item of the described items in a journal book file 43. The system is provided with a voice recognizing and inputting mechanism 10 having a registration file 35 in which the character string indicating the prescribed item is registered in advance. The mechanism 10 inputs the prescribed item displayed on the screen 41 in the form of voices, recognizes the voices, and produces a recognized character string and the registered character string to be allocated to the recognized character string is retrieved from the registration file. The retrieved character string is collated with the recognized character string and decided as the character string for the prescribed item on the screen 41. Then the prescribed item and the code number of the slip 12 are stored in the journal book file 43.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to an accounting system. In more detail, among items described in a voucher such as a sales voucher or a journal voucher (hereinafter simply referred to as a voucher), account items and numerical values are input into a so-called book file (hereinafter referred to as a journal book file) by voice input. The present invention relates to an accounting system in which an account item is automatically coded according to the input, and the account item can be automatically input.

[0002]

2. Description of the Related Art Conventionally, out of daily business slip processing, which is a daily business, and accounting processing necessary for tallying at the end of the month and closing at the end of the fiscal year, the items described in the slip are entered in a cash accounting book or a journal ledger. There is an accounting business that performs accounting processing of these book data.

In recent years, personal computers (personal computers)
Accounting management programs that can be carried out using such methods have already been put into practical use.
Normally, data such as amounts, accounts, items, etc., described in a slip are classified for each account, and the necessary data can be posted to various ledgers by inputting the amounts and the like into a personal computer using a keyboard or the like. . And
If this input and other tasks are performed, a considerable portion of the various outputs, such as tabulation calculations and list data, have been mechanized, making it much more convenient than before.

[0004]

However, of these accounting tasks, the task of inputting data such as account items and amounts by using a keyboard while viewing both the slip and the personal computer screen disperses attention to both. In addition, since the operation must be performed accurately as a whole, the operation is unexpectedly nervous and requires a heavy burden on the operator (operator (user)).

The present invention has been made in view of the above problems, and has as its object to provide an accounting processing system having an input system capable of minimizing the use of a keyboard and greatly reducing the burden of an input operation by an operator. And

[0006]

In order to solve the above-mentioned problems, an accounting processing system according to claim 1 of the present invention optically reads the items described in a slip 12 and displays them on a screen 41 of a display means. In an accounting processing system for collating the display contents of the described items and storing predetermined items of the described items in the journal book file 43, the accounting processing system may be a registration file in which a registered character string representing the predetermined items is registered in advance.
The voice recognition input mechanism 10 having 35, the predetermined items displayed on the screen 41 are input by voice and recognized as a recognized character string by voice recognition, and a registered character string to be assigned to the recognized character string is The registered file 35 is searched and compared as a candidate character string on the screen 41 and determined as a character string corresponding to the predetermined item. Further, when there is a code number corresponding to the predetermined item, the code number is assigned and the predetermined item is assigned. And the code number is stored in the journal book file 43.

[0007] With this configuration, a predetermined item can be voice-inputted without using a keyboard or the like, so that the attention of the input operation may be directed only to the confirmation of the described item.

[0008] In this specification, speech recognition means automatically extracting information on the meaning and personality contained in the speech (wave) using a computer or the like, and automatically extracting the meaning of the speech. To recognize.

According to the accounting processing system of the second aspect of the present invention, the voice recognition input mechanism 10 includes a voice input means 20 for obtaining a digital signal representing the voice waveform when the predetermined item is voice-input; The digital signal is recognized as a recognized character string by voice recognition, and a registered character string to be assigned to the recognized character string is searched from the registered file 35 and displayed as a candidate character string on the screen 41 for collation. And the voice recognition means 30 that determines the character string to be stored in the journal book file 43, so that highly accurate voice input can be performed.

According to the accounting processing system according to the third aspect of the present invention, the account name and the numerical value can be used as the predetermined items. The book file 43 can be reasonably created to be suitable for customary accounting operations.

According to the accounting system according to the fourth aspect of the present invention, the items described on the slip 12 can be read by the optical character reader 50, so that the journal 12 can be read only by placing the slip 12 on the reading section. Items necessary for creating the ledger file 43 can be easily displayed and converted into data.

[0012]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 shows the overall view of the accounting processing system 1 with exaggerated classification of each function, so that the voice recognition means 30 of the voice recognition input mechanism 10 is a conceptual block. FIG. 2 is related to FIG.
The flow of data and the like is also shown as a block for each function with some considerations.

The accounting processing system according to the present invention is intended for slips and the like that require accounting processing, and roughly includes a voice recognition input mechanism 10, an optical character reading device 50, and a display as shown in FIG. Central processing unit (hereinafter referred to as CPU
40).

First, the voice recognition input mechanism 10 is composed of voice input means 20 and voice recognition means 30. The voice input means 20 allows the operator to input the character string data (hereinafter, referred to as character string data) of the "word" representing the account name, the amount of money, etc., written on the slip in the form of a voice wave by the voice input 11. The sound wave is input to a processing system and converted into an analog electric signal.

The voice recognition means 30 further converts the analog electric signal obtained by converting the voice wave into digital data (hereinafter, referred to as a digital voice signal) by A / D conversion.
After recognizing the digital voice signal to obtain a recognized character string, a search is performed to assign registered character string data such as an account name registered in advance in the registration file 35 to the recognized character string, and a candidate character string is obtained. indicate. Then, the operator checks the candidate character strings, determines whether the character strings are correct or not, and determines the character strings for the predetermined items. Hereinafter, each will be described in more detail.

First, the voice input means 20 comprises a well-known microphone 21 and the like, and a voice board 22. Then, a speaker (speaker) who is an operator utters the word as a slip description as a sound wave, inputs the sound to the microphone 21 or the like, and converts the sound into an analog electric signal representing a sound waveform. Convert. Thereafter, the analog electric signal is taken into a later-described voice recognition engine main body 31 via the voice board 22. At this time,
The audio board 22 is configured to perform a Fourier transform or the like on the audio waveform obtained by the microphone 21 to convert the audio waveform into a digital audio signal. For this reason, the audio board 22 has an arithmetic circuit or the like for the Fourier transform or the AD conversion. It is configured. The microphone 21 may be a telephone or the like instead.

Next, the speech recognition engine main body 31, which is the core of the speech recognition means 30, obtains characteristic parameters from the spectrum of the speaker's voice obtained as digital data, and evaluates them based on the wave height and the like for each frequency band. Recognition by performing speech recognition on character strings represented by digital speech signals, such as analyzing the pitch and frequency components of speech under certain conditions, such as the temporal change and resonance state of speech. This is a device having a function of obtaining a recognition character string as a result.
And, as will be described in detail later, the speech recognition engine body
The means around 31 collates with the registered character strings registered in the registration file 35 in advance to determine whether they match, and determines whether the character string is correct or not.

As a method of such a speech recognition, a DP for obtaining a matching path of a standard pattern with respect to an input pattern is used.
(Dynamic Programming) A matching method or a so-called Markov model is created in advance, and when an input speech spectrum is observed, the HMM (Hidden Markov
Model) method or the like is used.

At this time, the speech recognition engine main body 31 uses a feature parameter that does not receive a change in acoustic features caused by the speaker, in order to use a speech from an unspecified speaker. Then, in the registration file, as a registered character string, character string data represented by phonetic symbols described below,
Both character string data of ordinary characters as written words are registered. Therefore, the main body of the speech recognition engine 31
Can recognize speech without specifying the speaker.

That is, such a speech recognition engine body 31
In the above, the recognition character string, the International Phonetic Association (Inter
National Phonetic Association (IPA) uses the International Phonetic Alphabet (IPA) to represent the phonetic symbol string (IPA character string). In addition to expressing as a `` symbol string '', it is also expressed as `` orthographic character string (or orthographic character string) '' mixed with kanji or kana that is usually used as a written word, and the two correspond one-to-one, and the registered character string in a separate registration file The voice recognition engine main body 31 can search the registered character string data for a character string to be assigned to the recognized character string data obtained as a result of voice recognition from the voice signal. It is.

In recent years, computer programs that can easily execute such functions by a personal computer or the like have been developed as general-purpose software, and those that can be put to practical use have been sold. Also in this embodiment, as general-purpose software,
For example, "Automatic Speech Recognition Software Developmen" by L & H Japan Ltd.
t Kit for Windows 95 ”can be used to construct a speech recognition engine. In addition, software by Dragon Corporation or the like may be used, but such general-purpose software is preferable because the operator of the accounting processing system does not need to be limited in that the speaker is not specified, that is, there is no speaker dependence. And according to the aspect as general-purpose software, a CP described later is used.
As an application program installed in the U-PC 40, it can be used as one stored in a storage medium such as a CD-ROM or a floppy disk.

Next, the voice recognition means 30 is provided with the engine communication means 32,
A recognition accuracy determination unit 33, a search unit 34, a registration file 35, and the like are provided.
It operates according to a program stored in a microcomputer or a chip IC or the like including a memory such as a ROM and a ROM, and constitutes a voice recognition means 30 together with a voice recognition engine main body 31.

First, the engine communication means 32 communicates data between the speech recognition engine main body 31 and the recognition accuracy judgment means 33 and the like by API (Application Program Interface).
e) Control according to the rules. That is, generally, an exchange occurs between an application program such as the voice recognition engine main body 31 and each unit around the voice recognition engine main body 31, and it becomes necessary to use basic functions of the operating system. Sometimes, it is an aggregate of commands and the like prepared so that commands can be called from here. This makes it possible to easily interface with peripheral units.

The recognition accuracy judging means 33 is a component of the program software attached to the speech recognition engine main body 31 from a functional viewpoint. That is, the voice recognition engine itself
31 normally generates a recognition character string that is a plurality of recognition results for an input voice signal, and adds a numerical value (probabilistic determination value) representing relative certainty to each of them. The recognition accuracy determination means 33 receives the probabilistic determination value and searches these values, and when the maximum value exceeds the predetermined value TH, the recognition accuracy determination means 33 determines the recognition result. After determining that the character string is correct, the recognized character string is displayed as a candidate character string (on the CPU / PC 40). still,
The setting of the predetermined value TH is changed in consideration of actual speech recognition accuracy and processing efficiency of speech recognition.

If these numerical values do not exceed the predetermined value TH, a plurality of recognition results are displayed as a list of candidate character strings as they are. Therefore, the recognition accuracy determining means 33
Is an aggregate of commands such as a predetermined value TH, an arithmetic instruction, a logical instruction, and a display instruction.

The registration file 35 is a file in which data such as the account name in the notation of the IPA character string and the orthographic character string, and a code number corresponding to each account name on a one-to-one basis are stored. Yes, RAM and ROM
And so on.

According to the present invention, the method of registering the account name includes registering the account name itself,
Only the initials of the account names are registered in correspondence with the list in which all other account names starting with the initials are enumerated, voice recognition of the initials is performed, and candidate character strings are displayed in the form of the list. It is possible to both determine the account name from the list and to use the account name according to the frequency of use of the account name.

Therefore, it is possible to register all the account names that are rarely used (several times a year), and to register all the accounts as they are, to reduce the storage capacity, reduce the search efficiency, and increase the search time. The disadvantage of this can be avoided.

When registering the latter list in the registration file 35, the list is used as a candidate character string by the CPU.
When displayed on the PC 40, as shown in FIG.
It is displayed for each registered account name and stored so that the account name can be selected.

[0030] Regarding the amount of money, basically,
Since data can be input by an optical character reader described later, characters that are insufficiently optically read are corrected one digit at a time, and "0" to "9" are registered. The reading method shall be unified in this system.

The search means 34 compares the recognized character string data of the speech recognition result with the registered character string data stored in the registration file 35 and searches for the registered character string from a functional viewpoint. See voice recognition engine body 31
Is a component of the program software associated with. Therefore, the search means 34 is constituted by a program having a search function such as a well-known binary search method or hash function method.

By configuring the voice recognition input mechanism 10 as described above, the present accounting processing system 1 can convert a high-precision voice input 11 into a voice input and voice recognition (hereinafter, referred to as a standard) within a storage capacity of a personal computer. (Referred to as voice recognition input), and the operator can simplify the operation procedure to the extent that the items described on the slip 12 can be read aloud according to the instructions displayed on the screen 41 of the CPU / PC 40. it can. In the voice recognition input, the processing speed when inputting interactively with the screen can be obtained as long as the processing speed is not delayed, so that the operability is good.

Further, the above-described voice recognition input mechanism 10 itself is used as program software on a CD-ROM.
The present accounting processing system can be widely used via a storage medium stored in a storage medium such as a disk or a floppy disk.

Next, an optical character reader (Optical Chara)
A cter Recognition (OCR) 50 is a device that optically reads the items described on the slip 12, converts them into character codes, and sends the data to the CPU / PC 40. As the OCR 50, one having a generally known configuration can be used. This type of device is preferable because it is very easy to handle since the items described on the slip 12 can be displayed on the screen 41 of the CPU / PC 40 simply by placing the slip 12 on a predetermined reading unit of the device.

In the accounting system 1, the OCR
Slight reading errors due to 50 (in some cases, reading errors may occur due to distortion, dirt, bleeding, or rubbing) can be easily corrected by voice recognition input. It is preferable in terms of equipment cost. Of course, the device for reading the items described on the slip 12 is not limited to the OCR 50, but may be an image scanner, a camera, or the like.

Finally, the CPU / PC 40 is a personal computer main body 42 as a central processing unit that controls the operations of the speech recognition input mechanism 10, the OCR 50 and the journal file 43, and makes the entire accounting system 1 function. At the same time, the apparatus has a storage medium for storing data such as the account item and the amount of money finally input by voice input 11 as a journal book file 43 necessary for accounting processing.

At the same time, the CPU / PC 40 has a screen 41 as a display means.
A user interface for displaying the description items of the slip 12 read by the CR 50 and the portions of the description items that require voice input or correction confirmation, and performing the voice input 11 itself in the same manner as a normal personal computer operation. Used as part of As such a display unit, a CRT, a liquid crystal display panel, or the like is used. Furthermore, CPU / PC40
Is provided with an input device such as a keyboard and a mouse for other operations.

At this time, if the CPU / PC 40 is constituted by a personal computer on which an operating system (OS), which has been rapidly spreading in recent years, is mounted, the voice recognition input mechanism 10 according to the present invention can be used regardless of its manufacturer or model. It is preferable because it can be used interchangeably.

As described above, the voice recognition input mechanism 10
Therefore, the operator can input a predetermined item of the items described on the slip 12 by voice without using a keyboard operation or the like through manual operation of the operator. Therefore, the operator can turn the line of sight to the slip 12 (the actual item) and the screen 41 to input the above-mentioned items by voice with less movement of the fingers. In short, since the movement of the "hand" is not required, the attention of the input operation may be allocated only to the confirmation of the described items, and an unexpected input error or the like at the time of typing with the keyboard can be eliminated.

In other words, according to the present voice recognition input mechanism 10, since the contents of the slip 12 are only "confirmed" visually on the screen 41, there is almost no chance of an input error. Since the correct data can be confirmed and the correct data can be input by voice, the input can be made very accurately. In this way, the accuracy of book entry can be improved and the speed can be increased.

Hereinafter, the operation of the accounting system 1 according to the present invention will be described based on an embodiment with reference to FIGS. In the actual operation, the input procedure is slightly different depending on each of the following cases (1) to (3). The cases (1) to (3) are (1) frequently used accounts, (2) numerical values, (3) rarely (some times a year, etc.). ) In the case of inputting the account to be used.

First, a starting operation common to each of the cases (1) to (3) will be described. First, the accounting processing system 1 is started by the CPU / PC 40, the slip 12 is placed on a reading section (not shown) of the optical character reader 50, and the reading section 12
Screen of the display means of CPU / PC 40
Display by 41. In this screen 41, the slip 12 is displayed, and among the items described in the slip 12, the account name and the like are posted (stored) in the journal book file 43 by voice input.
Indication indicating the place of the prescribed matter to be done is made
(Described below). Then, an instruction such as a next item to be input is displayed on the screen 41 by the OS in a window format or by using an icon, so that the operator can easily operate the accounting processing system 1 according to the instruction.

First, the case of (1) inputting frequently used account items will be described. FIG. 3 is a flowchart showing from the voice input 11 to the determination of the character string data. In FIG. 4, the slip 12 is displayed on the screen 41, and among the items described in the slip 12, a portion 11a that requires voice input is shown as one of frequently used accounts.

In response to the instruction (not shown) displayed on the screen 41, the operator inputs the voice 11 by the voice input means 20 (step 82). The voice signal data is recognized by the voice recognition engine main body 31 (step 82). 83), recognition character string data is obtained (step 84). Then, it is compared with the registered character string, and after the recognition accuracy determination regarding the recognized character string data is performed (step 85), when the probabilistic determination value of the recognition accuracy determination result exceeds the predetermined value TH, the audio signal The character string data 14a is shown on the screen 41 as a candidate character string for.

For example, according to FIG. 4, of the account items,
The character string data 11a of "product sales" is shown in reverse display or shaded display, for example, and it is shown on the screen 41 that voice input is to be performed, so the operator reads "product sales" aloud. When input from the microphone 21, the voice signal is obtained in the voice input means 20, and the voice recognition means 30 generates the candidate character string data 14 to be assigned to the voice signal by using the data of “product sales” of the recognized character string.
14a and other candidates 14a 'are searched, and these are displayed on the screen 41.

In this case, since the data 14a is correct as compared with the character string data registered in the registration file 35, the result is selected (step 86). That is, at this time, since the character string of the data 14a of “commodity sales” is recognized as the correct account item name registered in the registration file 35 (the operator sees the screen 41 to determine and confirm).
A code number (not shown) to be assigned to the account name on a one-to-one basis is automatically obtained from the registration file 35 at the same time and stored in the journal book file 43 (step 87).

Next, the case (2) of inputting a numerical value of money will be described. In the following description, only points different from the case (1) will be described. FIG. 5 shows that the slip 12 is displayed on the screen 41, and among the items described in the slip 41, a portion 11 b that requires voice input
Is shown. According to FIG. 5, originally, the amount is “1800
It can be confirmed on the screen 41 that the second digit character that should become "00" could not be read by the optical reading in the OCR 50.

[0048] The screen 41 indicates, as "?", That this part 11b is a part to be corrected by voice input, so that the operator reads out "8" and inputs it from the microphone 21. The voice recognition means 30 uses the voice signal as data of a recognized character string "8"
b is displayed on the screen 41 as a candidate character string. in this case,
Other candidates were not shown because the recognition accuracy determination result was high. In this case, too, the data 14b is correct, so this result is selected.

Next, a description will be given of the case of inputting an account which is rarely used (eg, several times a year) in (3). In this case, for example, an account item that is rarely used, such as “capital”, is registered as a word together with another account item whose name starts with “shi”. Therefore, when there is a voice recognition input of "shi", FIG.
As shown in the list 14c registered in this "shi"
Is displayed on the screen 41 as the candidate character string data 14. With this display, the operator selects the corresponding account from the list 14c and fixes it as the account name.

The case where the recognition result is correct has been described above.
It is not always the case that the candidate character string of the recognition result displayed on the screen 41 is correct. Conversely, if the data of the recognition result is incorrect, the voice input is tried again. Hereinafter, the second and third rows of the form are similarly input by voice.

As described above, the correct account name and the numerical value are determined as the recognition result, and the account name and the numerical value and the code number corresponding to the account are obtained. Is stored in the journal book file 43. That is, the character string of the data 14a "commodity sales" is used as a correct account name together with the code number to be assigned to the account name.
Stored in 43 (step 88). Then, the screen 41 displays the recognition result
15 is displayed as the determined result (step 89).

As described above, the account item and the amount of money from the slip 12 are transferred to the journal book file 43, and the accounting process of the accounting system is completed. After that, the journal book file 43 is sent as data to a conventional general accounting processing program, and the conventional accounting processing is performed, whereby a totaling process and an output can be performed.

As described above, in the present embodiment, an example is described in which the account item and the amount are input by voice, and the code number corresponding to the account item is automatically assigned. However, the present invention is not limited to this example. Instead, if the storage capacity of the registration file 35 is increased and the storage data is increased, the slip data itself such as other items and dates can be input by voice.

At this time, increasing the input of items etc.
Although technically possible, of course, as the number of registered proper nouns increases, the storage capacity of the registration file will be squeezed, and at the same time, according to current accounting practices, tasks that can be satisfied with accounts and amounts will be sufficient. Since there are many, as in the present embodiment, the first method was to realize voice recognition of essential data as a journal book file with high precision and to create the journal book file 43 rationally to reduce the burden of many tasks. .

[0055]

The accounting processing system according to the present invention includes a voice recognition input mechanism, and the predetermined items displayed on the screen are input by voice and recognized as a recognized character string, and are assigned to the recognized character string. A registered character string to be performed is searched from the registered file, collated on the screen as a candidate character string, and determined as a character string for the predetermined item, and
When there is a code number corresponding to the predetermined item, the code number is assigned, and the predetermined item and the code number are stored in the journal book file. All you need to do is dedicate your attention to entering tasks,
Input errors due to typing can be eliminated.
Therefore, the accuracy of book entry can be improved and the speed can be increased.

Further, the voice recognition input mechanism includes a voice input means and a digital signal representing a voice waveform is voice-recognized into a recognized character string, and a registered character string to be assigned to the recognized character string is searched from a registered file. And a voice recognition means for displaying the text on the screen as a candidate character string and collating the same to determine the character string corresponding to the predetermined item and storing the character string in the journal book file. The operation procedure can be simplified within the capacity range.

Further, since the account item name and the numerical value can be used as the predetermined items, the data necessary as the journal book file can be recognized by speech with high accuracy, and the journal book file can be rationalized so as to be suitable for customary accounting operations. Can be created. Then, the accounting processing operation efficiency can be further improved by combining with conventional accounting processing software.

Further, since the entries on the slip can be read by the optical character reader, the entries necessary for creating a journal book file can be easily converted into display data simply by placing the slip on the reading section. In addition, it is possible to obtain the accounting software of the present invention while reducing the apparatus cost.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an accounting processing system according to an embodiment of the present invention, divided into device units.

FIG. 2 is a block diagram showing the configuration of the accounting processing system according to the embodiment of the present invention, which is functionally divided including the flow of data.

FIG. 3 is a flowchart showing a main part of an operation procedure of the accounting system according to the embodiment of the present invention;

FIG. 4 is a display screen diagram typically showing a part of the operation of the accounting system according to the embodiment of the present invention.

FIG. 5 is a display screen diagram typically showing a part of the operation of the accounting system according to the embodiment of the present invention.

FIG. 6 is an explanatory diagram of a part of a display screen showing an example of a list of candidate character strings in the accounting processing system according to the embodiment of the present invention;

[Explanation of symbols]

10 ... voice recognition input mechanism, 11 ... voice input, 12 ... slip, 20 ...
Voice input means, 30 voice recognition means, 31 voice recognition engine main body, 35 registration file, 40 CPU / PC, 43 journal entry file, 50 optical character reading device

Claims (4)

[Claims] 1. An item read on a slip (12) is optically read and displayed on a screen (41) of a display means. After comparing the displayed contents of the item, a predetermined item of the item is entered into a journal book file. (43) In the accounting processing system stored in (43), the accounting processing system comprises a speech recognition input mechanism (10) having a registration file (35) in which a registered character string representing the predetermined item is registered in advance, The predetermined item displayed on the screen (41) is input by voice and is recognized as a recognized character string by voice recognition.A registered character string to be assigned to the recognized character string is searched from the registered file (35) and is displayed as a candidate character string on the screen. (41) The character string corresponding to the predetermined item is determined by collation on the above, and when there is a code number corresponding to the predetermined item, the code number is assigned, and the predetermined item and the code number are registered in the journal book file ( 43) Accounting system characterized by being delivered. 2. The speech recognition input mechanism (10) comprising: speech input means (20) for obtaining a digital signal representing the speech waveform when the predetermined item is inputted by speech; and speech recognition of the digital signal. A registered character string to be assigned to the recognized character string is retrieved from the registered file (35), displayed as a candidate character string on the screen (41), and collated to determine the character string for the predetermined item. 2. The accounting processing system according to claim 1, further comprising a voice recognition means (30) for storing the data in a journal book file (43). 3. The accounting system according to claim 2, wherein the predetermined items are an account name and a numerical value. 4. The accounting system according to claim 1, wherein the items described in the slip are read by an optical character reader (50) and displayed on a screen (41).