NVDA Roadmap
This document outlines NV Access’s planned development roadmap for NVDA and its supporting infrastructure for 2026. Priorities are grouped into short, medium, and long-term time frames.
Short-Term Priorities
These priorities focus on enhancements and improvements that we aim to deliver in the near future.
- 64-bit migration: Finalising the update of NVDA’s low-level code to run natively in a 64-bit environment.
- Secure add-on runtime: A runtime execution environment, separate from NVDA, that allows untrusted add-ons to run with limited access to the user’s computer. The first version of the runtime will provide support for speech synthesis and braille devices, with more capabilities being added over time.
- On-device image description: Private, low latency image description running locally on the user’s computer. This is intended to complement the existing image description capabilities of add-ons for massive, cloud-only description models. On-device description provides short text descriptions (e.g. “a man holding a hat”) without any data leaving the user’s computer.
- MathCAT integration: Full integration of the MathCAT math expression reader directly into NVDA.
- Chinese word segmentation: Improved word segmentation capabilities for Chinese language texts.
- Reducing testing friction: A suite of changes designed to encourage community beta testing, including making error beeps optional and improving issue reporting documentation that is more easily accessible from within NVDA.
- Add-on store infrastructure: Improving performance and maintainability.
- Excel UIA support: Continuing to improve UI Automation support in Microsoft Excel for better performance and stability.
- Braille improvements: Focusing on braille stability and continuing support for multi-line braille displays.
Medium-Term Priorities
These priorities represent more significant features and improvements that require more extensive development effort, or that we aim to deliver after our short-term priorities.
- Magnifier: Launching the initial version of a built-in screen magnification feature to better support low-vision users, alongside specific UX improvements for mouse users.
- OCR improvements: Updates to on-device OCR capabilities, including allowing users to choose specific OCR models.
- Microsoft Natural Voices: Adding support for Microsoft’s Natural Voices text-to-speech engine.
- Microsoft Office UIA: Expanding UIA support beyond Excel to include other Office components, specifically scoping improvements for PowerPoint and Word.
- Remote access E2E encryption: Implementation of End-to-End (E2E) encryption for NVDA Remote.
- Educational outreach: Launching “Programming 101,” a curated learning portal to help blind/VI students transition from NVDA users to developers.
- Analytics architecture: Rebuilding our internal analytics collection system for scalability and cost efficiency, and implementing an add-on analytics pipeline to understand add-on usage patterns.
- Infrastructure modernisation: Replacing the NSIS installer and replacing py2exe to modernize the build and installation process.
- ARIA compliance: The annual push to maintain NVDA’s adherence to modern ARIA standards.
Long-Term Priorities
These priorities represent ambitious goals and initiatives that will shape the future of NVDA, or that we aim to deliver after our medium-term priorities.
- Magnifier enhancements: Iterating on the magnifier feature to include advanced cursor tracking and bug fixes.
- Corporate mode: Compatibility and security settings to assist the deployment and maintenance of NVDA in enterprise environments.
- Feature usage statistics: Anonymously tracking specific feature usage within NVDA to identify settings that are no-longer used.
- Braille font attributes: Utilising multi-line braille to intuitively communicate font attributes of text.
- Video call readiness: Audio cues on framing, lighting and obstructions for video calls.
- Document authoring assistance: Tools to help B/VI users identify accessibility issues (font choices, formatting) in documents they author.
- Cross-device configuration sync: Allowing users to have portable cloud-based configuration profiles shared across devices.
- UI element recognition: Leveraging machine learning to identify and interpret inaccessible UI elements.
Contributing and Feedback
We encourage contributions from the community!
Please explore NVDA’s GitHub repository or user groups to learn more about how you can get involved.