WO2006120193A1

WO2006120193A1 - Method and apparatus for entering characters into a data processing system

Info

Publication number: WO2006120193A1
Application number: PCT/EP2006/062158
Authority: WO
Inventors: Henry Feil; Ewald Frensch
Original assignee: Siemens Aktiengesellschaft
Priority date: 2005-05-10
Filing date: 2006-05-09
Publication date: 2006-11-16
Also published as: DE102005021526A1

Abstract

Disclosed are a method and an apparatus for entering characters into a data processing system (MFG4). According to the invention, a plurality of handwritten elements, each of which represents a set encompassing at least one character, are successively detected with the aid of a sensor device (GS). The respective sets of characters are then determined from the detected handwritten elements by means of a recognition device (EKE) so as to form a series of signs based on the determined sets of characters in the order in which the handwritten elements assigned to the sets have been detected. An acoustic output device (LS) then outputs the determined sets of characters in the order in which the handwritten elements that are assigned thereto have been detected. Said output is monitored by a user, who issues an acoustic corrective signal when a set of characters has been entered or recognized incorrectly, said corrective signal being received by an acoustic input device (MIK). An insertion device (EFE) finally inserts another set of characters in place of the determined set of characters in the series of signs whose acoustic output was followed by the reception of the corrective signal.

Description

description

Method and device for entering characters in a data processing system

The present invention relates to a method and a device for inputting characters in a data processing system, as well as a data processing system, in particular ^¬ special in the execution of a portable electronic Gerrät, which has a device for inputting characters.

The miniaturization of portable electronic devices, such as cell phones or organizers, involves a reduction in controls, and sometimes results in a total relinquishment of mechanical keys or keyboards. However, there is then more and more the problem of entering differentiated control statements, such as characters in a font input. In the case of larger portable devices, conventionally a touch-sensitive display, a so-called "touch screen" is provided on which characters or characters can be written by means of a special pen, which are then recognized by means of a special character recognition software as characters or control instructions. In smaller portable electronic devices, such as pen-shaped mobile phones (so-called "pen phones"), which usually have a small display, on the other hand, a user's movement is detected by the portable electronic device for inputting characters or control instructions. Specifically, holding a Benut ^¬ zer, designed as a pin mobile phone in his hand, he will move for inputting characters, the mobile phone or its tip as a conventional pen to describe by the movement of the desired character to be inputted. The detection of the movement can be done for example by means of optical sensors, such as in a Com ^¬ puter mouse, or by means of acceleration sensors. In Either way, it is also necessary here that the appropriate movements are recognized as characters with a spe ^¬ cial handwriting recognition software.

In each of the cases just described, however, it happens that in principle, the input of writing methods has the problem that erroneous entries or misrecognitions of the characters may occur. To remedy this, an efficient and ergonomic correction procedure is necessary.

A conventional correction method is to delete the input characters of a character string from the last character to the erroneous character by means of a correction key and then to add further newly entered characters to the position in the character string to which the deletion has been performed. This correction method has ^¬ We sentlichen the disadvantage that a user already ne ^¬ admit and correct by the handwriting recognition software recognized character again has to re-enter. In the case of font input systems in which input characters are acoustically output for verification by the user after the input or recognition - somewhat delayed due to the calculation in the recognition - then the user must remember exactly which of the characters of several input characters typed incorrectly or had been recognized.

Another way to correct faultily give ^¬ ner characters is to use a so-called navigation mode. It is assumed that a user has first set a so-called input mode to input multiple characters. For example, if the user detects that one of the typed characters has been misrecognized, he must change from input mode to navigation mode. This can be done for example by pressing a special mode button. Subsequently, the user must move a cursor to the location of the misrecognition by a corresponding input. Thereupon it is necessary to switch back to the input mode to delete and correct incorrectly recognized characters. Finally, to return to the end of the entered sequence of characters, it is necessary to switch back to navigation mode, and once there, the input mode has to be activated in order to input new characters at the end of the character string and to add them to the character string. Thus shows that a correction of a character string in accordance with the very laborious methods just described and unkomforta ^¬ bel for the user.

The object of the present invention is thus to provide a possibility for inputting characters, in which also a simple correction is possible.

This object is solved by the independent claims. Advantageous embodiments are the subject of Unteransprü ^¬ che.

In this case, a method for entering characters or characters into a data processing system comprises the following steps. It is recognized in a plurality of segments Handschriftele ^¬ succession, one of which consisting of at least each represents a set of a character. A handwriting element can be understood as a movement performed by a user to describe a character. On the one hand, the movement currently being performed by the user to describe one or more characters can be detected, but it is also conceivable to record already recorded movements for the description of one or more characters. In the latter case this can be a user by means of color of a forth ^¬ conventional pen on a paper or write multiple characters, the paper then dium serves as Aufzeichnungsme- for movement to the description of characters, and can by means of an optical scanner then the recorded movement of the user as the character or characters are captured. As already mentioned, the Set of characters include one or more characters, ie single letters, numbers or whole words. The respective sets of characters are then determined from the recorded handwriting elements in order to form a character string in the order of detection of the handwriting elements assigned to the sentences on the basis of the ascertained sets of characters. The determination of the respective characters or sets of characters takes place here by means of special character recognition methods, which are usually realized by means of special software. In such recognition methods, a detected character is compared with a predefined character, in which case the most predefined character to the detected character is output as the recognized character. If the sets of characters have been determined, these are output acoustically in a next step in the order in which the handwriting elements assigned to them are recorded. A user of the very phrases entered by characters representing handwriting elements pursuing the acoustic output of the determined sets of Schriftzei ^¬ chen and are, if it detects a false entry or false detection, an acoustic correction signal. For example, this acoustic correction signal may include a noise, such as a whistle, but may also include a special correction word such as "FALSE." This acoustic Cor ^¬ rektursignal will be received. Subsequently, a ^further set of characters is inserted in the place of the ascertained set of characters in the character string, to the acoustic output of which the acoustic correction signal was received. That is, in the method according to the present invention, the user who detects a wrong input or misrecognition of a character does not need to remember exactly when the wrong input or misrecognition has taken place, but only has to if he perceives the acoustic output of an incorrect character to give off its acoustic correction signal. This simplifies the correction of the input character string and thus increases the convenience of inputting characters. According to an advantageous embodiment of the invention, the further set of characters for insertion into the string is provided such that after receiving the acoustic correction signal another handwriting element of a user is detected, which represents the further set consisting of at least one character. This white ^¬ tere set of characters is then again, as described above, identified and inserted in the place of the replaced set of characters.

According to a further embodiment of the invention, the further set of characters is provided in such a way that, after receiving the acoustic correction signal, a speech element of a user which comprises a word representing the further set of characters is detected. That is, according to this embodiment of the invention, no movement of a user but the user's speech is detected. Then the other set of characters is based on the word of the language element ermit ^¬ telt. In this case, in the determination, the word can be processed as the smallest linguistic unit, ie consisting of one or more letters, but it is also possible to recognize the meaning of the word and instead, for example, in the case where the word describes a number the verbal description of the digit directly to regard the numeral as the further set of characters to be inserted in place of a set of characters to be corrected in the string.

As already mentioned, capturing handwriting elements may be accomplished by detecting movement of a user with which the user describes a character. The detection of a motion of a user can thereby follow for example by means of an acceleration sensor, an optical sensor and / or a touch sensor ^¬ it. If an acceleration sensor is used to detect the movement of a user, then the acceleration sensor measures the accelerations occurring in the movement for describing characters, which can finally be converted by means of temporal integration into a locus or movement curve of the user in the description of characters. This movement ^curve can then be handed over to a character recognition method in order to recognize a character from it.

An optical sensor for detecting a movement of a user may comprise, for example a camera with wel ^¬ cher in equidistant time steps images of ground (a plane on which the movement for the description of handwriting elements takes place) are recorded and are compared with each previous images (striking Pixels representing the same background objects are compared between a current and a previous image), which is referred to as "pixel matching". Motion vectors can then be calculated from the difference images, which can be integrated to the new position finally, thus a locus or motion ^¬ the movement curve of a user in the description to get out of character. The movement curve is then fed back to a text recognition process. Furthermore, it is possible already to a recording medium such as a sheet of paper to record written handwriting elements by means of an optical scanner and from this sign the font ^¬ be seen.

There are essentially two different possibilities when using touch sensors. Firstly, it is possible that can be designed as a touch-sensitive display, particularly when using the method in electronic devices having a larger display, the display, wherein a user by means of a pin ^¬ font characters can be described on the touch-sensitive display. Furthermore, it is possible that when using the In smaller portable electronic devices, for example, a touch sensor may be provided on a tip of a portable electronic device configured as a pen-shaped mobile phone, the sensor comprising, for example, a rotatable ball at the tip which rotates as it moves over a contacting surface , And based on the rotation on the movement of the Ge ^¬ device, which is guided by a user, can be closed.

In accordance with another aspect of the invention, an apparatus for inputting characters into a data processing system is provided. In this case, the device comprises a sensor device for detecting a plurality of handwriting elements in succession, each of which represents a sentence comprising at least one character. Furthermore, the device comprises a recognition device for determining the respective sets of characters from the recorded handwriting elements in order to form a character sequence in the sequence of the detection of the handwriting elements assigned to the sentences on the basis of the ascertained sets of characters. In addition, an acoustic output device is provided for acoustically outputting the determined sets of characters in the order of detection of the handwriting elements assigned to them. Finally, an inserter is used to insert another set of characters in the place of the ascertained set of characters in the string to whose acoustic output the correction signal has been received.

According to an advantageous embodiment, the sensor device may comprise an acceleration sensor, an optical sensor and / or a touch sensor. For an explanation of the respective sensors, reference is made to the illustration with regard to the method.

According to a further advantageous embodiment, the acoustic output device comprises a loudspeaker and comprises the acoustic input device a microphone. The insertion device and / or the recognition device can each have a microprocessor. However, it is also conceivable that the insertion device and / or the recognition device are designed as software applications which are executed by a microprocessor.

It is conceivable that the device further comprises a memory device for storing the character string. This makes it possible to enter a string completely, to store in the storage device and then to transfer at once, for example via a data cable to the data processing system.

Further, the apparatus may include display means for displaying the input character string. This allows the user to perceive the input characters not only acoustically (by the acoustic output device) but also optically.

According to another aspect of the invention, there is provided a data processing system comprising a character writing device just described. The data processing system can be designed as a portable electronic device. In particular, it can be designed as a mobile device or mobile telephone, but also as an organizer or PDA (PDA: Personal Digital Assistant).

Preferred execution of the present invention advertising the following with reference to the accompanying Zeichnun ^¬ gen explained in more detail. Show it:

FIG. 1 shows a time structure in which the individual event lines for inputting characters are represented;

Figure 2 is a schematic representation of the correction of a misrecognized character; FIG. 3 shows a time structure with event lines for explaining the input of sets of characters or words;

Figure 4 is a schematic representation for correcting a misrecognized word;

5 shows a data processing system in the form of a mobile phone according to a first embodiment of the

Input of characters;

6 shows a data processing system in the form of a Mo ^¬ biltelefons according to a second execution form for entering characters;

7 shows a data processing system in the form of a Mo ^¬ biltelefons according to a third From guide die for inputting characters;

8 shows a data processing system in the form of a Mo ^¬ biltelefons according to a fourth From guide die for inputting characters.

Before a method for inputting characters will now be explained with reference to FIGS. 1 to 4, a possible device for inputting characters will first be explained with reference to FIGS. 5 to 8.

It should first be made to Figure 5, in which a Mobilte ^¬ LePhone MFGl shown. The mobile phone MFGl here has ei ^¬ ne conventional shape in the shape of a cuboid. It has an acoustic output device with a loudspeaker LS, as well as an acoustic input device with a microphone MIK. Furthermore, it has a keyboard TAS with multiple buttons for entering characters or control statements. Below the loudspeaker LS, a display device DSP is arranged on ^¬ , having a touch-sensitive screen. This touch-sensitive screen provides thereby a Be ^¬ rührungssensor BS. With a pen STI, it is then possible to descendants of characters, while the tip of the pin STI with the display DSP is in contact handwriting elements representing ^¬ sign one or more font to enter. In this case, a user can, for example, the tip of the pin STI to the Anzeigeein ^¬ direction DSP put on, and perform a movement which corresponds to the number "1", wherein the value detected by the touch sensor BS movement trajectory of the pen tip is ^¬ forwarded to a Erken drying apparatus EKE, the it is directed ^¬ is, from the detected movements of the pen tip, that is the signature elements, each characters or sets of characters to identify. The recognition device also serves to classify recognized characters or sets of characters into a sequence of characters, namely in the order of detection of the handwriting elements assigned to the characters. As will be explained later with reference to FIGS. 1 to 4, an insertion device EFE serves to insert one or more characters at a specific position to be corrected into the character string generated by the recognition device EKE.

Referring now to Figure 6, there is shown an apparatus for entering characters in the form of a pen MFG2. The mobile telephone MFG2 essentially comprises the same components with the same functions as the mobile telephone MFG1, namely an acoustic output device with a loudspeaker LS, a display device DSP, a recognition device EKE, an insertion device EFE, and an acoustic input device with a microphone MIK. Due to the design of the mobile phone MFG2 as a pen, the number of keys for operating the mobile phone is greatly reduced. Thus example ^¬ example in Figure 6 only a keyboard Tasl shown with two keys. As a further difference to the mobile phone MFGl to ^¬ summarizes the mobile phone MFG2 instead of the touch sensor BS an optical sensor OS, which is provided in the tip section of the mobile phone ^¬ . The optical sensor OS, including an op ^¬ diagram camera, is used in equidistant time ^¬ take steps images of the surface OF, and to compare each case preceding recorded images in order to derive motion vectors of the tip of the mobile phone MFG2, which extends over the Surface OF moves, to calculate. At ^¬ the motion vectors eventually hand can position Points calculate which can then be combined into a movement curve or locus. The rule so by the optical sensor OS ^¬ detected movement curve of the tip of the Mo ^¬ biltelefons MFG2 is then again supplied to the EKE identification device that determines based on the movement curve the BEWE ^¬ supply curves associated characters. The function of the insertion device EFE will be explained later with reference to FIGS. 1 to 4.

Referring now to Figure 7, there is shown another apparatus for inputting characters in the form of a mobile phone MFG3 formed as a pen. The mobile phone MFG3 corresponds in structure essentially to the mobile phone MFG2, for which reason reference is made to the explanation of the same components on the explanation of the mobile phone MFG2. The main difference of the mobile phone MFG3 is now that it has a touch sensor BS1 instead of the optical sensor OS in the mobile phone MFG2. This can for example have a rotatably mounted Ku ^¬ gel at the top of the mobile phone MFG3, the rotation of which is then measurable when the ball or the touch sensor BSl is moved on contact with the surface OF on this way ^¬ away. Based on the detected rotation, the movement of the tip of the mobile telephone via the MFG3 Oberflä ^¬ is then che away detectable. Is described with the tip of a character, it can detect the touch sensor BS that convert into a movement curve or locus and these Be ^¬ wegungskurve the recognizer EKE out. As already explained with reference to the other embodiments, the Recognition device EKE then able to recognize the respective character using the movement curve.

Finally, reference is made to FIG. 8, in which a further device for inputting characters in the form of a mobile phone MFG4 designed as a pen is shown. In this case, the mobile telephone MFG4 again essentially matches the mobile telephones MFG2 and MFG3, for which reason reference is made to the above-described mobile telephones in order to explain the same components. Characteristic of the mobile phone MFG4 is that it now has a gyro sensor or acceleration sensor GS in the tip section of the phone to detect movements. This gyro sensor has at least two sensor sections, but advantageously three sensor sections for detecting accelerations in two, advantageously in three dimensions. That is, the Mo ^¬ biltelefon MFG4 is or its tip portion, in which the acceleration sensor GS is, moved over the surface OF, for example, to write a character to be sawn, the acceleration sensor GS detects at the top of the mobile phone MFG4 occurring accelerations. By double integration of acceleration data on the time a motion curve or locus leaves the BEWE ^¬ supply the top of the mobile phone MFG4 calculated. This locus is then in turn fed to the detection device EKE, which determines a character or several characters from the movement curve.

After the basic explanation of the structure of a possible device for capturing characters, a method for capturing characters or for their correction in the event of erroneous detection or incorrect determination (by the recognition device) will now be explained with reference to FIGS.

A method for inputting individual characters in the form of numerals will first be described with reference to FIGS. 1 and 2. In Figure 1 are four event lines shown in which the time from left to right vorreirei ^¬ Tet. The first event line, the "font input" event line, indicates that the digits "0", "1", "2", "3", "4" are inputted successively by a user at respective time intervals TSZ. The input can be effected by means of one of the mobile phones MFG1 to MFG4 shown in FIGS. 5 to 8. More specifically, in the font ^¬ a respective manuscript element HSE input by a user by means of a respective sensor (BS, OS, or BSl GS) ER- sums. As already mentioned, in the example of FIG. 1, a handwriting element HSE comprises a character SZ in the form of a number.

The detected movement of a user to describe a digit or a handwriting element is then fed to a respective detecting means EKE, which determines the the handwriting element underlying paragraph by means of a ^¬ be known handwriting recognition method. This requires a certain computing time, which is why the second Ereignisli- never that "recognized input" -Ereignislinie, time nachfol ^¬ quietly begins. In the example, assume that he ^¬ are five known characters or digits SPC, with four digits of the digit sequence input IF (see FIG. 2) are correctly identified, while the third number "2" is incorrectly recognized by the recognize voltage device so that the third digit now represents a "4" instead of a "2".

Finally, in the third event line, the "speech output" event, the characters ESZ recognized by the recognition device are detected by the acoustic output device or its loudspeaker LS, which is referred to herein as a so-called "text-to-speech" system Language "system) work, issued in the form of words EWO. Based on the calculation for the speech output, the acoustic output of the recognized characters ESZ begins with a time delay to the detection or determination of the respective characters. It is assumed that the output of EWO words that he knew ^¬ represent characters or numerals, in a uniform speed occurs. This means that from ^¬ reproducing a respective digits is done at equal time intervals TAK.

Referring now to Figure 1 of the first event line to the third event line so noticeable that due to between ^¬ time calculations and the acoustic output at a constant speed, for example, with He ^¬ version of the fourth digit has already been started, while using acoustic speech output just the recognized third digit is output. In order to avoid that a user who recognizes from the acoustic voice output that his third entered digit ("2") has been detected incorrectly, the just entered fourth digit ("3") must clear again or navigate cumbersome to the misrecognized digit in order to finally arrive at the wrongly entered third digit as provided in conventional character correction systems, the present invention provides a simplified approach according to this embodiment.

According to this embodiment of the invention, a user who is currently inputting characters only has to keep track of the acoustic output of the recognized digits, and immediately when he recognizes that a digit has been entered incorrectly or misrecognized immediately after the output of that digit (generally said after output of the character) output an acoustic correction signal, for example in the form of the word "FALSE". As shown in Figure 1 with respect to the fourth event line, the "Speech Entry" event line, a user recognizing at the Speech Output Event Line that the third character entered by him was misrecognized immediately outputs the word "FALSE" , By this acoustic correction signal KOS in the form of the word "FALSE" after a false sign FZ the actual correction process is started. The individual correction steps will now be explained in detail with reference to FIG. 2, in which four screen sections D1, D2, D3 and D4 are shown, which in each case represent sections of the display device DSP which follow one another temporally during the correction process. In Bildschirmab ^¬ cut Dl while the recognized according to the second event line of Figure 1 string ZF or numeric string "01434" is displayed with the cursor CUR indicates the end of the sequence of digits that a user subsequently to the recognized digits additional digits or characters, can enter. During the optical representation of the recognized digit sequence, the digit sequence is likewise emitted acoustically, which is illustrated by the loudspeaker symbol to the left of the screen section D 1 and has already been explained with reference to the third event line of FIG. If now the third recognized digit is output as the digit "4", then the acoustic correction signal of the user sounds in the form of the word "FALSE", so that the digit just emitted acoustically is marked as false. The acoustic correction signal is received by an acoustic input device of a mobile telephone or its microphone MIK. The acoustic input device serves as an integrated voice recognizer, which can be operated, for example, in the so-called "voice activation" mode, ie in which the voice recognizer is always active and triggers a corresponding predetermined action upon recognition of a specific word.

Such a marking of an incorrect character or a wrong digit can, as shown in the second screen section D2, again be effected by a cursor CUR.

It is possible in this case for the entire correction method, which is initiated by the output of the acoustic correction signal by the user, to be controlled by a correction device in the form of the insertion device EFE designated in FIGS. 5 to 8. In this case, after receiving the acoustic correction signal, the string of the recognized characters ESZ and the body of the Schriftzei ^¬ chens are communicated to the output of the akusti ^¬ cal correction signal was received. The inserter then causes, as already explained with respect to screen section D2, that the cursor CUR is now set to the wrong one

Character FZ or in this case the wrong digit FZ is set. The marking now means that the marked character is replaced by a subsequent input of a character or a number.

This is illustrated in the screen section D3. An entry of a new character can take place on the one hand by detecting a handwriting element HSE, which represents a character SZ, as has already been explained with respect to the first event line of FIG. On the other hand, the input of a new character can also be done by voice input by the user outputs a voice element SPE, which includes a word WO, which represents the character or a number. In the present case, the correct number "2" should be inserted instead of the wrongly recognized number "4". As can thus be seen in the third screen section D3, in which he now correct number, which is entered upon receipt of the acoustic correction signal, inserted in place of the misrecognized digit ^¬ knew digit sequence.

As shown in the screen section D4, after inserting the correct character (s), the cursor is returned to the end of the digit string so that a user can now add new characters at the end of the digit string. That is, the navigation of the cursor from the end of the digit sequence to the incorrectly entered digit to correct it, and the navigation from that position back to the end to the digit string after correction is performed automatically (here, in the example of the EFE inserter). Thus, a simple method is available to the user without much effort to correct incorrectly entered characters. Reference is now made to Figs. 3 and 4 which illustrate a method for entering and correcting whole words, each consisting of a set of characters. In the present case, it is thus assumed that a user not only inputs individual digits, but entire Tex ^¬ te with consisting of letters words. The input of the respective letters for the words can be carried out according to the input of the numbers, as has already been explained with reference to FIG. In this case, according to the first Er ^¬ eventislinie of Figure 1 by detecting the movement of a user who represents a letter, a font ^¬ input or handwriting input. More specifically ^¬ NEN individual handwriting elements of a user Kgs detects the advertising comprising a set consisting of several characters. To identify the input device or the recognition device and insertion device, which characters belong to a specific handwriting element, a user can press, for example, after entering a handwriting element one of the keys of the keyboards TAS or TASl. It is also conceivable that the user at the beginning of entering a handwriting element presses one of the keys of the keyboards TAS or TASl and keeps pressed until the input of the handwriting element is completed. In this way it is thus possible, individually recorded

Characters, as explained according to the first event line of Figure 1, respectively, then, as it has been explained according to the second event line of Figure 1 ^¬ recognized by the recognition device EKE by means of a handwriting recognition method, and then as words or Combine sets of characters in a string. Such a string of recognized words or characters is finally shown in Figure 3 in the form of a text TXT, which, as has also been explained with respect to Figure 2 can be displayed on the display of one of the mobile phones MFGL to MFG4 after detection. To explain the correction method according to a further embodiment of the invention, reference is now made to the first event line of FIG. 3, which substantially corresponds to the third event line of FIG. The words just recognized by the recognition device EKE are audibly output by the acoustic output device or its loudspeakers LS to a user for checking purposes. As can already be seen in the representation of the text TXT, an error has occurred in the recognition of the fourth word "spaer". The user now hears the individual words EWO issued and recognizes in the fourth word that this has been recognized incorrectly. Thus, immediately following the output of the fourth (false) word FW, it outputs an acoustic correction signal KOS in the form of the word "FALSE," as shown in the second event line of FIG.

The acoustic correction signal KOS is received by the microphone MIK of the acoustic input device (one of the mobile phones MFG1 to MFG4), whereby finally the actual correction process is started. This correction process will now be explained in a simplified manner with reference to FIG. According to the second screen section D2 of FIG. 2, for example, in the screen section D5 of FIG. 4 again controlled by the insertion device EFE (one of the mobile phones MFGl to MFG4), a cursor CUR is set to the erroneous word. This means that the word marked by the cursor CUR should be replaced by a subsequent input of a word or a set of characters. In the present case, by means of a font input - symbolized on the left side of FIG. 4 - a hand ^¬ writing element HSE of a user is detected, which represents the correct word or the correct set of characters SSZ. After detecting this correction word, the word (more precisely, the individual characters) is recognized by the recognition device EKE and inserted in the text TXT instead of the incorrectly recognized word. This can be seen in the image ^¬ screen section D6 of Figure 4. Subsequently, each subsequent input of characters or sentences of Characters appended to the end of the text TXT. This also means in the case illustrated in FIGS. 3 and 4 the replacement of a wrongly entered or incorrectly recognized word is carried out conveniently, without a user having to navigate from the end of an entered text to the incorrect position, and having to navigate back to the end after a correction ,

Claims

claims

1. A method for entering characters in a data processing system, comprising the following steps:

- Detecting a plurality of handwriting elements (HSE) in succession, each of which represents a sentence consisting of at least one character (SZ);

- Determining the respective sets of characters from the detected handwriting elements (HSE) to form a string (ZF; TXT) based on the determined sets of characters (ESZ) in the order of detection of the handwriting elements assigned to the sentences;

- Acoustic output of the determined sets of characters (EWO) in the order of detecting the handwriting elements associated therewith;

- Receiving an acoustic correction signal (KOS) of a user;

- Inserting another set of characters (SSZ, WO) in the place of the determined set of Schriftzei ^¬ chen (FZ, FW) in the string on the acoustic output towards the acoustic correction signal wur ^¬ de received.

2. The method of claim 1, wherein after receiving the acoustic correction signal (KOS) another handwriting element (HSE) is detected, which represents the further set consisting of at least one Schriftzei ^¬ chen, and the further set of characters is determined.

3. The method according to claim 1, wherein after receiving the acoustic correction signal (KOS) a speech element (SPE) is detected comprising a word (WO) representing the further set of the character and the further set of characters is determined.

4. The method according to any one of claims 1 to 3, wherein the detection of handwriting elements (HSE) comprises detecting a movement of a user, with which the user describes a character (SZ).

5. The method of claim 4, wherein detecting a movement of a user by means of an acceleration sensor (GS), an optical sensor (OS) or a touch sensor (BS, BSl) takes place.

The method of any one of claims 1 to 3, wherein detecting handwriting elements (HSE) comprises scanning written characters.

7. A device for entering characters in a data processing system, having the following features:

a sensor device (BS, BS1, OS, GS) for detecting a plurality of handwriting elements (HSE) in succession, each of which represents a set comprising at least one character;

- A recognition device (EKE) for determining the respective sets of characters (SZ, SSZ) from the detected handwriting elements (HSE) to the basis of the determined sets of characters a string (TXT) in the order of detection of the sentences to form zugeord ^¬ Neten handwriting elements;

- An acoustic output device (LS) for the acoustic output of the determined sets of characters in the Sequence of detecting the handwriting elements assigned to them;

an acoustic input device (MIK) for receiving an acoustic correction signal (KOS) of a user;

- An inserter (EFE) for inserting another set of characters in the place of the determined set of characters (FZ; FW) in the string on the acoustic output to the correction signal (KOS) has been received.

8. Apparatus according to claim 7, wherein the sensor device comprises an acceleration sensor (GS), an optical sensor (OS) or a touch sensor (BS, BSl).

9. Device according to claim 7 or 8, wherein the acoustic output device comprises a loudspeaker (LS) and the acoustic input device comprises a microphone (MIK).

10. Device according to one of claims 7 to 9, wherein the insertion device (EFE) and / or the recognition device (EKE) have a microprocessor.

The apparatus of any one of claims 7 to 10, further comprising memory means for storing the string.

12. Device according to one of claims 7 to 11, further comprising a display device (DSP) for displaying the character string (TXT).

13. Data processing system (MFG1, MFG2, MFG3, MFG4) with a device for inputting characters according to one of claims 7 to 12.

14. Data processing system according to claim 13, which is designed as a portable electronic device.

15. Data processing system according to claim 14, which is designed as a mobile radio device (MFGl, MFG2, MFG3, MFG4) or a PDA.