*NVDA Addon Guide: Vision Assistant Pro* Vision Assistant Pro is an advanced, multi-modal AI assistant for NVDA. It leverages world-class AI engines to provide intelligent screen reading, translation, voice dictation, and document analysis directly through your screen reader. Below is the official guide to installing, configuring, and using the add-on.
*Step 1: Installation* * Open your NVDA Menu (NVDA + N). * Navigate to Tools > Add-on Store. * Search for Vision Assistant Pro, select install, and restart NVDA when prompted. *Step 2: How to Get a Free Gemini API Key* Since this add-on requires an AI provider, it is highly recommended to use Google Gemini for the best performance and accuracy. If you do not have an API key, you can create one for free by following these quick steps: * Go to the Google AI Studio website (aistudio.google.com). * Sign in with your standard Google/Gmail account. * Click on the "Get API key" button (usually located at the top left or main dashboard). * Click "Create API key", select your project or create a new one, and generate the key. * Copy the long string of text and numbers provided. Keep this safe, as you will need to paste it into NVDA. *Step 3: Setup & Configuration* To configure the add-on, go to NVDA Menu > Preferences > Settings > Vision Assistant Pro. 1.1 Connection Settings * Provider: Select Google Gemini (strongly recommended for the best performance with images and files). * API Key: Paste the API key you copied from Google AI Studio here. * Fetch Models: After entering your API key, press this button to download the latest list of available models from your provider. * AI Model: Select the main model you want to use for general chat and analysis. *1.2 Advanced Model Routing (Optional)* Check "Advanced Model Routing (Task-specific)" if you want to assign specific models to different tasks (e.g., a specialized Vision model for images, or a specific Speech-to-Text model for dictation). ⚠️ Warning: Choosing an incompatible model for a task will cause errors. *1.3 General Preferences* * OCR Engine: Choose Chrome (Fast) for quick text extraction, or AI (Advanced) for superior layout preservation. * TTS Voice & Creativity: Select your preferred AI voice style and adjust the "Creativity (Temperature)" slider. Lower values are better for accurate translations and OCR. *Step 4: Command Layer & Shortcuts* * To prevent keyboard conflicts with NVDA or your applications, this add-on uses a Command Layer. * Press NVDA + Shift + V (the Master Key) to activate the layer. You will hear a confirmation beep. * Release those keys, then press one of the following single keys to execute a command: * O — Full Screen Vision: Analyzes the entire screen layout and content. * V — Object Vision: Describes the current navigator object. * D — Document Reader: Advanced reader for PDFs and images with page range selection. * C — CAPTCHA Solver: Captures and solves CAPTCHAs (supports Gov portals). * S — Smart Dictation: Converts speech to text. Press to start recording, press again to stop and type. * T — Smart Translator: Translates text under the navigator cursor or selection. * E — UI Explorer: Identifies and clicks UI elements in any application. * Shift + A — AI Operator: Autonomous Operation. Tell the AI to perform a task on your screen. * Space — Recall Last Result: Shows the last AI response in a chat dialog for review or follow-up. * H — Commands Help: Displays a list of all available shortcuts. *Document Reader Shortcuts (Inside the Viewer)* When using the Document Reader (D), you can navigate and manage pages using these specific shortcuts: * Ctrl + PageDown / PageUp: Move to the next or previous page. * Alt + A: Open a chat dialog to ask specific questions about the document. * Alt + R: Force a re-scan with your active AI provider. * Alt + S or Ctrl + S: Save the extracted text as a .txt or .html file. *Please Note* An active internet connection is required for all AI features to function. -- With regards, [image: NAB Delhi has campus in R.K. Puram, Dwarka and Narela in Delhi] <https://www.nabdelhi.in/> *Helpline for the Blind* *National Association for the Blind* Sector 5, R.K. Puram, New Delhi 110022 India W: www.nabdelhi.in E: [email protected] T: +91 8826261166 M: +91 9212319672 [image: Connect wiht us, we are on social media] <https://ngo.nabdelhi.in/QR_page/NAB_QR1.html> [image: Donate] <http://www.nabdelhi.in/donate> *Empowered over 1,30,000 visually impaired since 1979* [image: Beyond Eyes Foundation, our social enterprise][image: NAB Delhi has received Best Institution National Award from Government of India] [image: NAB Delhi certified for good governnance by Credibility Alliance] *80G | FCRA | CSR Preferred Partner | Tax exemption in USA 501(c) (3) * Disclaimer: This email and any attachments transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error, please notify the sender immediately and delete the email from your system. Any unauthorized use, disclosure, or distribution of this email is strictly prohibited. NAB Delhi makes every effort to keep its email communications free from viruses. However, we cannot guarantee that this email or any attachments are free of viruses or other malicious code. It is the recipient's responsibility to ensure appropriate measures are taken to scan for viruses. -- -- Disclaimer: 1. Contents of the mails, factual, or otherwise, reflect the thinking of the person sending the mail and AI in no way relates itself to its veracity; 2. AI cannot be held liable for any commission/omission based on the mails sent through this mailing list.. Search for old postings at: http://www.mail-archive.com/[email protected]/ --- You received this message because you are subscribed to the Google Groups "AccessIndia" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/a/accessindia.org.in/d/msgid/accessindia/CADonsavRFga2ud_bs3zph260Wok3Wsd-fMwjp%3D4OQC4-P34GRA%40mail.gmail.com.
