Captcha Solver is an add-on for NVDA screen reader that provides a convenient way to recognize image captcha through the rucaptcha.com service. You can always get the latest version of Captcha Solver on this page

Customization and use
After installing the add-on, follow these steps:
  1. Register on the website rucaptcha.com and top up the balance of the created account in any convenient way.
  2. Copy your 32-digit API key from the page [ https://rucaptcha.com/setting ].
  3. Open the menu NVDA / Service / Captcha Solver / Settings and in the settings dialog box that appears, in the "API Key" field, insert the key obtained in the previous step and click the OK button.
  4. Open the NVDA / Options / Input Gestures menu and in the "Captcha Solver" section assign any free gesture (key combination, touch gesture, etc.) to the "Start Recognition Process" command. It is also recommended to assign a gesture to the "Report Account Balance" command to control expenses.
Now, using the NVDA object navigation, install the navigator on the object with the captcha image (as a rule, it is an object of the "graphics" type) and start the recognition process with the previously configured gesture. After the message about the successful sending of the captcha image, you should wait for the recognition result for 10-15 seconds, and if everything goes well, NVDA will inform you about it, and the recognized text will appear on the clipboard. The add-on is designed to recognize classic captcha with one line of characters in the image. Cyrillic captchas are also recognized without problems.

Additional settings
  • The selected checkbox "Recognize only graphic objects" blocks the recognition of a navigator object that is not a graphic element. This setting protects the user from attempts to recognize foreign objects, since the captcha object in most cases is a graphic element.
  • The checkbox "Match case when recognizing" in the CaptchaSolver settings dialog determines whether the employee solving the captcha should enter the answer taking into account the case of characters in the image. In most cases, the case of characters does not matter when responding to a captcha.
  • The checkbox "Inform image size" enables and disables reading the dimensions of the navigator object with a single call of the command to start solving the captcha. If this check box is selected, then the recognition command should be called twice quickly to start the recognition process. Sizing information can make it easier to find the desired CAPTCHA object.
  • The "Send text instruction" checkbox indicates whether the employee should send a text instruction to help solve the captcha. If the checkbox is checked, then before sending the captcha image, the user will be shown a dialog where he can enter the desired instruction text.
  • The Image Language combo box defines the language of the symbols in the image. For example, if you are solving a Cyrillic captcha, then in this list you should select "Cyrillic only", so that the captcha is guaranteed to get to the solution of a Russian-speaking employee.
Documentation https://githubmemory.com/repo/kvark128/captchaSolver