Looking for the latest version? Download Capture2Text_v3.5.zip (37.4 MB)
Home
Name Modified Size Downloads / Week Status
Totals: 2 Items   26.8 kB 9
Capture2Text 2014-07-18 1,313 weekly downloads
readme.txt 2014-07-18 26.8 kB 99 weekly downloads
Capture2Text Readme -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- What is Capture2Text? -------------------------------------------------------------------------------- Capture2Text enables users to do the following: * Optical Character Recognition (OCR) Allows the user to quickly snapshot a small portion of the screen, OCR it and (by default) save the result to the clipboard. * Speech Recognition (experimental) Using speech recognition the user can speak into their microphone and Capture2Text will convert the speech to text. If the speech recognition technology is not 100% sure, Capture2Text will present the user with a list of the most likely transcriptions. The selected result will (by default) be copied to the clipboard. -------------------------------------------------------------------------------- How to Launch Capture2Text (no installation required): -------------------------------------------------------------------------------- 1) Unzip the contents of the zip file. Make sure that there are no Asian or other non-ASCII characters in the path where you unzipped it. Also, if you are on Windows 7, don't unzip it to the Program Files directory (this will avoid issues related to write privileges). 2) Double-click on Capture2Text.exe. You should see the Capture2Text icon on the bottom-right of your screen (though it might be hidden in which case you will have to click on the "Show hidden icons" arrow). -------------------------------------------------------------------------------- OCR Language Support: -------------------------------------------------------------------------------- Capture2Text can OCR the following languages: Afrikaans Finnish Maltese Albanian Frankish Norwegian Ancient Greek French Polish Arabic Galician Portuguese Azerbaijani German Romanian Basque Greek Russian Belarusian Hebrew Serbian Bengali Hindi Slovakian Bulgarian Hungarian Slovenian Catalan Icelandic Spanish Cherokee Indonesian Swahili Chinese (Simplified) Italian Swedish Chinese (Traditional) Japanese Tagalog Croatian Kannada Tamil Czech Korean Telugu Danish Latvian Thai Dutch Lithuanian Turkish English Macedonian Ukrainian Esperanto Malay Vietnamese Estonian Malayalam By default only English, French, German, Italian, Japanese and Spanish are installed. To acquire other languages: 1) Download the appropriate OCR language dictionaries from: http://code.google.com/p/tesseract-ocr/downloads/list These files end in ".tar.gz" (ex. tesseract-ocr-3.02.rus.tar.gz). 2) Open the ".tar.gz" file you just downloaded with 7-Zip or similar decompression software and navigate to the directory that has the file that ends in ".traineddata". 3) Drag the ".traineddata" file (and any other file in this directory) to this path in the Capture2Text directory: Capture2Text\Utils\tesseract\tessdata 4) Restart Capture2Text Note: Arabic and Hindi are more CPU intensive and will thus be slower to OCR. To install the Chinese (NHocr) language pack: 1) Download nhocr-0.18-dic-zh_CN-091226.tar.gz from https://code.google.com/p/nhocr/downloads/list 2) Open it with 7-Zip and copy PLM-zh_CN.dic and cctable-zh_CN to Capture2Text\Utils\nhocr\Dic 3) Restart Capture2Text -------------------------------------------------------------------------------- OCR Usage: -------------------------------------------------------------------------------- Press the OCR capture key (default: Windows Key + Q) to start the capture. Now, using your mouse, resize the blue capture box over the area of the screen that you want to OCR. A preview of the captured OCR'd text will appear in the top-left corner of the screen shortly after you have stopped resizing the capture box. Press the capture key again or the left mouse button to complete the capture. The captured screen area will be OCR'd and the textual result will be stored in the clipboard by default. To cancel an OCR capture, press ESC. To move the capture box, hold down the right mouse button and drag the mouse. To nudge the capture box, use the arrow keys. To toggle the active capture box corner, press the space bar. To change the OCR language, right-click the Capture2Text tray icon, select the OCR Language option and then select the desired language. To quickly switch between 3 languages, use the OCR language quick access keys: Windows Key + 1, Windows Key + 2 and Windows Key + 3. When the Tesseract versions of Chinese or Japanese is selected, you should specify the text direction (vertical/horizontal/auto) using the text direction key: Windows Key + W. The text direction will not have any effect on the NHocr Chinese/Japanese dictionaries. If auto is selected, horizontal will be used when the capture width is more than twice the height, otherwise vertical will be used. The text direction also affects how furigana is stripped from Japanese text. (For Japanese) When OCR pre-processing is enabled, by default, Capture2Text will attempt to strip out furigana. You may disable this behavior in "Preferences... -> OCR -> Strip Furigana". Using the Preferences dialog, you can change the following OCR settings: - OCR Hotkeys. - Current OCR Language. - The 3 Quick-Access OCR Languages. - Capture Box color and opacity. - Enable/Disable the preview box and change its colors, font and opacity. - Change the text direction (used for Chinese and Japanese). - Specify whitelist characters. - Enable/Disable OCR pre-processing. - Enable/Disable stripping of furigana (for Japanese). -------------------------------------------------------------------------------- Speech Recognition Language Support: -------------------------------------------------------------------------------- Capture2Text can perform speech recognition for the following languages: Afrikaans French Polish Chinese German Portuguese Czech Italian Russian Dutch Japanese Spanish English Korean Turkish Note: The speech recognition feature is still experimental and may not work properly. -------------------------------------------------------------------------------- Speech Recognition Usage: -------------------------------------------------------------------------------- Press the speech recognition capture key (default: Windows Key + A) to start the capture. You will see a box that says "Recording..." in the top-left corner of your screen. Speak a word or phrase or sentence into your microphone. Capture2Text will automatically recognize when you are done speaking and will display a box that says "Analyzing...". The speech recognition will take a couple of seconds. When the speech recognition is complete you will see a list of possible transcriptions to choose from. When you choose a transcription, it will be stored in the clipboard by default. When the results windows is displayed, you can press Enter to select the first transcription or use the number keys (1-9) to select the corresponding transcription. To cancel a speech recognition capture, press ESC. To change the speech recognition language, right-click the Capture2Text tray icon, select the Speech Recognition Language option and then select the desired language. To quickly toggle between 2 languages, use the speech recognition language hotkey: Windows Key + 4 Using the Preferences dialog, you can change the following speech recognition settings: - Speech recognition Hotkeys. - Current speech recognition Language. - The 2 speech recognition languages to toggle between. - The properties of the Results window (font, color, number of results). - How much silence to wait for before recording stops. -------------------------------------------------------------------------------- Output Options: -------------------------------------------------------------------------------- By default, the OCR'd or speech recognized text will be placed in the clipboard. You also have 3 more ways to output the text. To send the text to a pop-up window you can right-click the Capture2Text tray icon and select Show Popup Window. To send the text to whichever textbox currently contains the blinking cursor/I-beam, right-click the Capture2Text tray icon and select Send to Cursor. (Advanced) To send the text directly to a window/control (for example, Notepad++), first fill in the Send to Control settings in the Preferences dialog. Using the Preferences dialog, you can change the following output settings: - Text to prepend/append to the captured text. - Enable/Disable outputting to the clipboard. - Enable/Disable outputting to a popup window. - Popup window properties (default width and height). - Enable/Disable sending the output text to the cursor. - Enable/Disable outputting to a control. - Additional command to send to the output control. -------------------------------------------------------------------------------- Configuration: -------------------------------------------------------------------------------- Right-click the Capture2Text tray icon in the bottom-right of your screen and then select the "Preferences..." option to bring up the Preferences dialog. -------------------------------------------------------------------------------- Substitutions: -------------------------------------------------------------------------------- Sometimes Capture2Text consistently makes the same OCR mistakes such as recognizing an "M" as "I\/|". By editing the substitutions.txt file in the Capture2Text directory, you may tell Capture2Text to substitute one text string for another text string. Just find the appropriate language section and add one substitution per line in this format: from_text = to_text Example (adding 3 substitutions to the English section): English: I\/| = M >< = X some%space%text = some_text To create a substitution regardless of language, add the substitution to the "All:" section. Special tokens and escape characters: %space% = Space character ( ) %tab% = Tab character ( ) %eq% = Equals (=) %perc% = Percent sign (%) %lf% = Linefeed character (\n) %cr% = Carriage return character (\r) You may disable a substitution by adding a "#" in front. When done editing substitutions.txt, either restart Capture2Text or switch language for the substitutions to take effect. -------------------------------------------------------------------------------- Command Line Options: -------------------------------------------------------------------------------- You may OCR the screen via command line by calling Capture2Text in this format: Capture2Text.exe x1 y1 x2 y2 [output_file] Required arguments: x1 - X1-Coordinate of the screen y1 - Y1-Coordinate of the screen x2 - X2-Coordinate of the screen y2 - Y2-Coordinate of the screen Optional arguments: output_file - The OCR'd text will be written to this file if specified. Capture2Text will read settings.ini to determine settings such as OCR language and output options (clipboard, popup, etc.). Examples: Capture2Text.exe 10 152 47 321 output.txt Capture2Text.exe 10 152 47 321 Notes: Make sure that you wait for Capture2Text to finish processing before attempting to start a new instance of Capture2Text. -------------------------------------------------------------------------------- Troubleshooting & FAQ: -------------------------------------------------------------------------------- Issue #1: Capture2Text doesn't work at all. What can I do? Possible solutions: - Make sure that you have unzipped Capture2Text to a path that does not contain Asian or other non-English characters. Search Google if you do not know how to unzip a file. - Make sure that you did not unzip Capture2Text to the Program Files directory. This will avoid possible issues related to write privileges. - Try unzipping Capture2Text to a very short path such as C:\Capture2Text. Some computers supposedly have issues with longer paths. I have never actually seen this happen though. - Make sure that your Anti-virus software is not blocking Capture2Text. Refer to the documentation that was bundled with your Anti-virus software. - Make sure that you have downloaded the latest version from SourceForge: http://sourceforge.net/projects/capture2text/files/Capture2Text/ - Restart your computer. - Right-click Capture2Text.exe and select "Run as administrator". Capture2Text shouldn't need administrator privileges, but if all else fails it won't hurt to try. - Ask one of your grandchildren to help you. If none of these things work for you, I suggest deleting Capture2Text and looking for some other OCR software (do not ask me for recommendations). Issue #2: Capture2Text stopped working. Just restart it. I have never actually seen this happen though. Issue #3: The language that I'm interested in doesn't appear in the OCR language menu. Read the OCR Language Support section of this document to learn how to add new languages. Issue #4: I don't see the Capture2Text tray icon. Perhaps Windows did you the disservice of hiding it for you. Click the "Show hidden icons" button (it looks like an triangle) to see the Capture2Text tray icon. Issue #5: I've clicked on the Capture2Text tray icon but it doesn't do anything. Right-click it instead. Issue #6: Capture2Text isn't properly OCR'ing some word or character. Since I don't maintain the OCR training files, there is nothing that I can do about it. If you have a technical background and a lot of free time, feel free to create your own training files: https://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3 Issue #7: Capture2Text isn't working on my Mac. Capture2Text is a Windows-only software. If you have a technical background, feel free to port it (but don't ask me to help). Issue #8: Speech recognition isn't working very well. Speech recognition is an experimental feature and I might remove it in a future release of Capture2Text. Issue #9: Where is the uninstaller? There isn't one. Capture2Text doesn't have an installer either. To remove Capture2Text from your computer, simply delete the Capture2Text directory. Issue #10: Should I use "Japanese (Tesseract)" or "Japanese (NHocr)"? "Japanese (Tesseract)" is usually more accurate and properly supports both horizontal and vertical text. If it makes a mistake, you can try using "Japanese (NHocr)". "Japanese (NHocr)" treats all text as horizontal text, so when using it to OCR vertical text, make sure that you only select a column of text that is one character in width. -------------------------------------------------------------------------------- Notes About the OCR Languages: -------------------------------------------------------------------------------- * In general - Handwritten text is not supported. - Smaller text is harder to OCR than larger text. - Text with an outline surrounding it is difficult to OCR. - Unless otherwise stated, all languages use the Tesseract OCR engine. * "Japanese (Tesseract)" - You may select the text direction: vertical/horizontal/auto. The default is "auto". - Usually more accurate than "Japanese (NHocr)". * "Japanese (NHocr)" - Uses the NHocr OCR engine. - Vertical text has a higher miss-rate than horizontal text. * "Chinese - Simplified (Tesseract)" and "Chinese - Traditional (Tesseract)" - You may select the text direction: vertical/horizontal/auto. The default is "auto". * "Chinese (NHocr)" - This uses the NHocr OCR engine. -------------------------------------------------------------------------------- Tips to Increase OCR Accuracy: -------------------------------------------------------------------------------- - Make sure that what you are viewing is >= its original size. That is, make sure that your image viewer isn't zoomed out too much. - Sometimes it helps to shape the capture box very tightly around the text. - Enable the OCR pre-preprocessing option in Preferences -> Output. -------------------------------------------------------------------------------- How to Run Capture2Text from Source Code: -------------------------------------------------------------------------------- 1) Download and install AutoHotkey 32-bit Unicode from: http://ahkscript.org/download/ 2) Copy the .ahk files from SourceCode/Capture2Text_AHK_Script to the root Capture2Text folder (the same folder that contains Capture2Text.exe) 3) Double-click Capture2Text.ahk. -------------------------------------------------------------------------------- Contact: -------------------------------------------------------------------------------- Christopher Brochtrup cb4960@gmail.com I do not provide technical support. Instead, see the Troubleshooting & FAQ section. I will accept feature suggestions, but no promises that I will actually implement your suggested features. -------------------------------------------------------------------------------- Acknowledgments/Tools Used: -------------------------------------------------------------------------------- Tesseract - OCR engine. NHocr - Alternate OCR engine for Japanese and Chinese. Leptonica - Image processing and analysis library. SoX - Audio utility used here to capturing voice to FLAC file. Google - Voice recognition service. wget - Utility used here to send voice recognition request to Google. ScreenCapture - AutoHotKey screen capture script. -------------------------------------------------------------------------------- Related Tools for Japanese Language Learners: -------------------------------------------------------------------------------- - JGlossator - http://jglossator.sourceforge.net/ Automatically lookup Japanese words that you have OCR'd with Capture2Text. Supports de-inflected expressions, readings, audio pronunciation, example sentences, pitch accent, word frequency, kanji information, and grammar analysis. Supports both EDICT and EPWING dictionaries. - OCR Manga Reader - https://play.google.com/store/apps/details?id=com.cb4960.ocrmr&hl=en Free and open source Manga reader android app that allows you to quickly OCR and lookup Japanese words in real-time. There are no ads and no mysterious network permissions. Supports both EDICT and EPWING dictionaries. -------------------------------------------------------------------------------- Version History: -------------------------------------------------------------------------------- [Version 3.5 (7-17-2014)] - Capture box should be less jumpy. - Preview will now only update when the user has stopped moving the capture box for at least 400 milliseconds. - When preview is setting to "Dynamic", the positioning should be less jumpy. [Version 3.4 (7-10-2014)] - Added option to strip furigana from Japanese text. - Added the "Auto" choice to the "Text direction" preference. - Removed the option to toggle "OCR pre-processing" from the Preferences. It may still be edited in settings.ini. - Changed the default "OCR pre-processing" hotkey to Shift-Ctrl-Windows-B. [Version 3.3 (3-2-2014)] - More minor tweaks to the Preferences dialog. [Version 3.2 (3-1-2014)] - Minor tweaks to the Preferences dialog. [Version 3.1 (2-28-2014)] - Improved OCR accuracy through use of better image pre-preprocessing (leptonica_util). - Now supports text and backgrounds of any color when OCR pre-processing is enabled. (In the previous version, only dark text on a light background was supported). - Added option to place the preview text beside the capture box. - Japanese (Tesseract) accuracy is now vastly improved through use of a Japanese-specific Tesseract config file. Also using this config file with Chinese (Tesseract). - Using Tesseract v3.02.02 for Japanese (was v3.01). - Replaced the binarize option with the OCR pre-processing option. - Removed "Send to Control" from the right-click menu. - Removed the Chinese (NHocr) language pack from default distribution. (You can still download it from https://code.google.com/p/nhocr/downloads/list). - Added the Italian language pack to the default distribution. - Removed setting of PreviewRemoveCaptureBox from the GUI. - Removed ConvertImageFormat (replaced with leptonica_util). - Now compiled with AutoHotkey 32-bit Unicode v1.1.14.03 (was v1.1.11.01). [Version 3.0 (8-27-2013)] - Added option to binarize captured image before sending it to the OCR engine. [Version 2.5 (7-5-2013)] - Updated NHocr from v0.20 to v0.21. - Now compiled with Ahk2Exe v1.1.11.01 instead of v1.1.05.06. [Version 2.4 (12-29-2012)] - Added support for Arabic, Danish (Alternate), Esperanto (Alternate), German (Alternate) and Slovakian (Alternate). [Version 2.3 (11-9-2012)] - Added option to remove the capture box before a preview OCR. This is more accurate, particularly with NHocr, but causes the capture box to flicker. - Changed the default image scale factor from 300% to 320% to meet Tesseract's minimum recommended DPI. - When using Japanese, revert to Tesseract v3.01. It is MUCH more accurate than v3.02.02. - Now passing a .ppm image to NHocr instead of a .pgm image to better handle non-grayscale captures. - Increased update rate of the capture box to make it appear more fluid. - Fixed text direction being ignored bug for Chinese/Japanese that was introduced in v2.2. - Fixed bug that caused the capture box to stick around after it was supposed to be removed. [Version 2.2 (11-4-2012)] - Upgraded to Tesseract v3.02.02. For details, see: http://code.google.com/p/tesseract-ocr/wiki/ReleaseNotes - Added whitelist option to the OCR tab. - Simplified substitution tokens and fixed whitespace bug. [Version 2.1 (10-7-2012)] - Added the substitutions feature. - Added command line options. [Version 2.0 (3-10-2012)] - Added the Preferences dialog. No more editing settings.ini by hand. - The popup window is now multi-lined. - Added option to preserve newline characters. - Limited preview to 150 characters. A trailing "..." will appear if necessary. - Added Speech Recognition Language option to right-click menu. - Cleaned up the right-click menu. - On the first run, inform user how to access the Preferences dialog. [Version 1.10a (2-18-2012)] - Removed GdiPlus.dll from distribution. [Version 1.10 (12-31-2011)] - Added preview box (and corresponding settings) [Version 1.09 (11-10-2011)] - Fixed speech recording stopping in the middle of a sentence. - Fixed VoiceMaxResults not working correctly. Also increased to 9 as default. [Version 1.08 (11-06-2011)] - Upgraded Tesseract to version 3.01 (it has better vertical text support and doesn't ignore small captures as much) - When using Tesseract Chinese or Japanese, you can now select the text direction (vertical or horizontal). To support this, added TextDirectionToggleKey and textDirection to settings.ini. - Changed default for ScaleFactor from 4.0 to 3.0 in settings.ini. - Changed menu text for Chinese and Japanese to reflect the OCR engine being used. [Version 1.07 (11-05-2011)] - Added voice recognition support via unofficial Google voice recognition service - Added the "Send To Cursor" option to menu. The setting.ini file includes: SendToCursor SendToCursorApplyBeforeAndAfterCommands - Renamed OCRAdjustment to OCRSpecific in settings.ini - Moved the CaptureBox section in settings.ini to the OCRSpecific section - Added VoiceSpecific to settings.ini. Section includes: VoiceMaxResults VoiceResultsWindowWidth VoiceResultsWindowFont VoiceResultsWindowFontSize VoiceSilenceBeforeStop VoiceLanguage - Added StartVoiceCapture to Hotkey section in settings.ini - Added VoiceLanguageToggleKey to Hotkey section in settings.ini - Removed scaleFilter from settings.ini - Removed the scaleFactor option from the menu (it's still in settings.ini) [Version 1.06 (12-12-2010)] - Added language quick access keys. - For Chinese and Japanese delete newlines. For other languages replace newlines with spaces. [Version 1.05 (12-04-2010)] - Fixed issue where the checkboxes in the language menu wouldn't disappear. [Version 1.04 (12-04-2010)] - Added ability to move the capture box by right-clicking - Added languages supported by the Tesseract OCR tool - Created a right-click menu that allow the user to select language, output type, capture box settings and scale factor - Removed unnecessary items from settings.ini [Version 1.03 (11-27-2010)] - Added ability to change dictionary when the Dictionary setting in settings.ini - Added Chinese dictionary [Version 1.02 (11-27-2010)] - Changed CaptureKey to StartAndEndCaptureKey in settings.ini - Added EndOnlyCaptureKey to settings.ini - Added ToggleActiveCaptureCornerKey to setting.ini [Version 1.01 (11-27-2010)] - Added ReplaceControlText to settings.ini - Added ability to use linefeeds, carriage returns and tabs in PrependText and AppendText - Added an "About" item to the tray menu. - Removed the capture box showing up in the taskbar - Removed the PassThruKey settings in settings.ini. They are no longer needed. - Changed the tray tooltip text - Cleaned up code and put the ScreenCapture routines in a separate file [Version 1.00 (11-26-2010)] - Initial version --------------------------------------------------------------------------------
Source: readme.txt, updated 2014-07-18