This article provides a comprehensive overview of prior updates and feature specifications for Windows Voice access.
Introducing fluid dictation in voice access
Fluid dictation is a new feature in voice access on Copilot + PCs, that makes voice-based dictation smoother and smarter. It automatically corrects grammar, punctuation, and filler words as you speak, reducing the need for manual editing. This experience is powered by on-device small language models (SLMs) so ensure fast and private processing. To use it, launch voice access and complete setup if you’re a first-time user. Fluid dictation is enabled by default – you can check or toggle it via the settings flyout in the top-right corner or simply say “turn on/off fluid dictation.”
It works in any text app but is disabled in secure fields like passwords or PINs to protect your privacy. Fluid dictation in voice access is currently available in all English locales and is available on all Copilot + PCs.
Wait time before acting setting in voice access
We are introducing a new ‘Wait time before acting’ setting in voice access, allowing users to configure the delay before a voice command is executed. This provides greater flexibility for users with varying speech patterns, enabling more accurate recognition whether speaking slowly or quickly. To set this up, navigate to Voice access settings> Wait time before acting, and choose the option that best fits your preferences.
Voice access in Japanese
Voice Access now supports Japanese, expanding accessibility for more users. You can now navigate, dictate, and interact with Windows using voice commands. With this update, Windows users who speak Japanese can enjoy a hands-free, voice-powered PC experience.
Extended support for flexible and natural commanding
We released more flexible and natural command execution in voice access on Copilot+ PCs. Voice access now understands multiple variations of an existing command. You can say:
-
“Can you open Edge application”
-
“Switch to Microsoft Edge”
-
“Please open the Edge browser”
And voice access will recognize your intent and execute the command accordingly. No more struggling to remember exact phrasing—just speak with flexibility!
Add custom words to the dictionary
We have introduced the ability for you to add your own words to the dictionary in voice access. Adding your own words, including difficult to pronounce words, to the dictionary in voice access will help improve dictation accuracy. It increases the probability of recognizing these words more accurately by creating a bias that helps voice access become more context aware and proactively understanding your specific vocabulary. The feature will be available in all the currently supported voice access languages.
You can add your own words to the dictionary in voice access in a few ways:
-
Directly from Voice access settings> Add to vocabulary.
-
After spelling a word using “Spell that” command or making a correction using “Correct that”
-
Anywhere when you want to manually add a word by using “Add to Vocabulary” command.
Discover what’s new on voice access with the latest update
We’re excited to share an immersive new way to stay up to date with voice access on Windows. With every update, voice access is getting smarter, faster, and more intuitive—and now, discovering these new capabilities is just as seamless as using them.
Right after launching voice access post major update, this new update window will appear highlighting the important features and improvements. You can review it right away or close it and come back to it later by clicking on the “What’s New” option in the setting menu on voice access bar.
From now on, you’ll never miss all the exciting new features coming along!
Improvements to the Spellings & Corrections Experience
We have made improvements by offering more and better suggestions for the Spellings & Corrections experience. Users will now be able to use these features throughout the entire VA session. The Spellings experience will enable users to dictate characters and whole words at a faster speed, providing more editing flexibility within the UI. This is achieved by supporting more dictation commands for selection, deletion, and text navigation.
Auto Restart for Voice Access
In Windows 11, we have introduced auto restart for voice access. For people with limited mobility, this feature makes sure that voice access will restart automatically in the event of an issue, allowing them to resume using voice access as soon as possible. To reactivate voice access, users won't need to depend on other assistive technology or help from third parties. We are making it simpler to report crashes that cause voice access to restart automatically so that the team may investigate into them and work on fixes.
New Voice Access Command for Windows Search
We are introducing a new command for searching directly with Windows search. Users can now say the following commands to search for their desired application or file in Windows ("Entity" is replaced with whatever application or file you want to search.)
-
Search "Entity",
-
Search Windows for "Entity",
-
Search for "Entity"
Voice shortcuts
You can now create your own commands using the voice shortcuts feature in voice access. These commands can be customized for single actions, such as opening a link, or multiple actions performed in sequence.
-
To start, activate voice access and say the command "Open voice shortcuts."
-
On the Voice shortcuts page, say Create new command button to initiate the command creation.
-
On the Create new shortcut page, dictate the name for the command, choose one or multiple actions, and enter the necessary information. For details, go to Use voice to create voice access shortcuts.
Note: This feature is only available in English–US, English–UK, English–India, English–New Zealand, English–Canada, and English–Australia.
Voice access support on multiple displays
You can now use all voice access features across multiple displays. Previously, some features like number and grid overlays were limited to the primary display. The grid overlay feature has been improved to allow you to interact with the display you're using and smoothly switch to another display when required. You can also use the drag & drop feature to move the files, apps, and documents between multiple displays.
-
To switch to another display, invoke the grid overlay and say the command "<Monitor>." Use the alphabet or NATO-phonetic identifier for the display you want to focus on. For example, saying "B" or "Bravo" would both be valid.
-
You can also use the quick command "Mouse grid <monitor> <number> <number>" to move your mouse to a specific point on a monitor. For example, say "Mousegrid A 114" to direct your pointer to a specific point on monitor A. To learn more about this feature, go to Use voice access on a multi-display setup.
Voice Access is now more intuitive and user-friendly
On Snapdragon-powered Copilot+ PCs, Voice Access now allows for more flexible and natural command execution. Usually, voice commands required a fixed syntax; if you didn’t use the exact phrasing, the command wouldn’t work.
For example, previously, you had to say “Open Edge” to launch the browser. If you said, “Can you open Edge application,” the command wouldn't be recognized.
With this update, Voice Access now understands multiple variations of an existing command. You can say:
-
“Can you open Edge application”
-
“Switch to Microsoft Edge”
-
“Please open the Edge browser”
Voice Access will recognize your intent and execute the command accordingly. No more struggling to remember exact phrasing—just speak with flexibility!
Additionally, if a command isn’t recognized, Voice Access will provide real-time command suggestions based on your last utterance, helping you refine your input.
Voice Access in Chinese
Voice Access now supports Chinese, expanding accessibility for more users. You can now navigate, dictate, and interact with Windows using voice commands in the following Chinese language variants:
-
zh-CN (Simplified Chinese - China)
-
zh-TW (Traditional Chinese - Taiwan)
With this update, Windows users who speak Chinese can enjoy a hands-free, voice-powered PC experience.
Voice access in global languages
Voice access is now available in additional languages, including Chinese, German, Spanish–Spain, Spanish–Mexico, French–France and French–Canada.
When voice access is turned on for the first time, you will encounter a prompt to download a speech model to enable on-device recognition of voice data. If voice access doesn't find a speech model matching your display language, you can still choose to proceed ahead to use voice access in English–US.
You can always switch to a different language by navigating to Settings>Language on the voice access bar.
Dictate text with voice access for Narrator users
Narrator, the built-in screen-reading application in Windows, now integrates with voice access. Use voice access to dictate text and listen to what you've dictated. You can also give commands to the Narrator using your voice, like "Turn on Narrator" or "Read selection."
To access the full list of Narrator commands supported with voice access, click the Help icon on the voice access bar and select "View all commands," then choose "Narrator commands."
Note: To learn more about Narrator, refer to the Complete guide to Narrator.
What works:
-
Starting and setting up voice access, including downloading a speech model.
-
Reading a list of voice access and Narrator commands from the voice access Help menu.
-
Changing the voice access microphone state.
-
Dictating text with voice and hearing back what was dictated.
What doesn't work:
-
Using voice access guide with a screen reader.
-
Core voice access features:
-
Using different voice access commands to get different tasks done.
-
Using "Click" commands or number overlays to interact with specific items on the screen.
-
Using grid overlays to control the mouse pointer.
-
Using text editing features such as corrections and spellings in voice access.
-
Tips:
-
Press Alt+Tab to cycle through open applications and focus on the voice access UI.
-
If you're not using headphones with the screen reader, the voice access microphone might pick up the audio output from the screen reader, causing unwanted behavior.
