***AI-generated Executive summary***
This transcript captures a Microsoft community webinar on new and improved Windows Narrator capabilities designed to help blind and low-vision users work more efficiently across Windows and Microsoft 365. The team frames the work as community-driven and then highlights five areas: more natural “HD” voices, significantly improved Word reading/editing/collaboration, AI-powered visual descriptions through Copilot, deeper personalization of what Narrator announces, and everyday features for control and privacy (speech recap, screen curtain, etc.). The Q&A focuses on ongoing Braille driver compatibility issues, automatic language switching requirements, and interest in additional synthesizer support, along with where to send feedback and stay updated.

Note: No editing of the summaries or transcripts was done.
***AI-Generated One-page structured summary***
Purpose: Share recent Narrator improvements, explain how the Windows Accessibility team gathers feedback and co-builds with the community, and demonstrate features that reduce friction in daily work (reading/writing, understanding visuals, and controlling verbosity/privacy).
Key improvements highlighted:
* Natural & Natural HD voices: More expressive pacing/emphasis; downloaded once, then available offline; offered in multiple languages and voices.
* Word productivity improvements (in partnership with Word team): More reliable continuous reading and editing; spoken confirmation for many formatting/actions; clearer table/list structure announcements; faster comment navigation and interaction (claimed reduction in keystrokes).
* AI descriptions for images and screens: On-demand descriptions via a shortcut that opens Copilot with the selected image/screen; follow-up questions supported; on Copilot+ PCs, initial descriptions can be produced locally before optionally escalating to Copilot.
* Personalization of announcements: A “personalize announcements” flow (demoed in Excel) to tune what Narrator reads (e.g., name/value only; whether a cell contains a formula), using either natural-language instructions or checkbox controls.
* Everyday essentials: Rehear/copy last phrase, Speech recap (recent Narrator output log), Braille viewer, Screen curtain (privacy), Speech off (avoid overlap with other audio), and a “What’s new” experience for recent updates.
Who it helps / scenarios: People who rely on a built-in Windows screen reader for work/school/home tasks—especially those reading long documents, editing and collaborating in Word, reviewing spreadsheets, navigating noisy web pages, or needing privacy in public spaces.
Notable shortcuts mentioned:
* Turn Narrator on/off: Ctrl + Windows + Enter
* Scan mode on/off: Narrator key (Caps Lock or Insert) + Space
* Describe image: Narrator + Ctrl + D (opens Copilot with the item for description and follow-ups)
* Screen description: Narrator + Ctrl + S (described as describing the full screen)
* Personalize announcements: Narrator + Alt + P
* Rehear last phrase: Narrator + X; Copy last phrase: Narrator + Ctrl + X
* Speech recap: Narrator + Alt + X
* Screen curtain toggle: Narrator + Ctrl + C
* Speech off toggle: Narrator + Shift + S
* Element lists on web pages: Narrator + F5/F6/F7 (landmarks/headings/links, as described)
Q&A themes: Braille display connection/driver conflicts (a filter-driver approach exists but issues remain for some device classes; the team hints at an upcoming update), automatic language switching (depends on correct language tags in the content and installed voices), and requests for broader synthesizer support. There is also discussion of desktop vs. web versions of Microsoft 365 apps and offline support for HD voices (offline after initial download).
Resources mentioned (as stated in the transcript): Contact Windows Accessibility; Windows accessibility resources; Disability Answer Desk / Enterprise Disability Answer Desk; Feedback Hub (including guidance to file Narrator-related issues and provide examples of pages/documents where language switching fails); and an “ask me anything” question-submission link shown during the webinar.

***AI-generated Transcript***
The following is a transcript of https://www.youtube.com/watch?v=SV0qfjG1Vs4, which was a webinar put on by Microsoft’s Narrator team. I am in no way associated with Narrator or Microsoft and produced this transcript for my own use from the YouTube recording via Microsoft Office 365 transcription.

Speaker 1
Hello, everyone, and welcome to the session today on getting more done with Narrator. Thank you for making time to join us today. My name is Ankur Khator. I lead the Windows Accessibility product team. And today, I'm not alone. I have my amazing team, three people who know Narrator better than me. are joining in here. Nidhi Jain and Roshni Padar, our product managers on the narrator team. And a lot of what you'll see today is something that they have been defining working on building. And then I also have Srihari Shirla, who is on the narrator engineering team, and he's the person who's doing the actual work, turning the feedback and all the amazing explorations into code. And just a quick word about the team and how we work, Windows Accessibility Team, Our mission is very simple. We want to make sure that Windows work for everyone, however you see, hear, move, or think. We work across segments. We build narrator for blind users. We build magnifier for low vision, voice access for mobility, and speech, live caption. All the assistive tools that you see in Windows are part of the things that we have been building. And also one more thing to admit, this is the first time that we are trying this format. It's a first for us. We have done a lot of customer conversations, advisory board, research calls, but an open community webinar like this is new for us. So we are very excited and a little bit nervous as well. We are very curious to see what works here for all of you and what doesn't. So feel free to tell us in the chat, tell us afterwards what's working, what's not working. And then anytime you have a question, please feel free to pop up in the chat. We'll take a pause. We'll certainly keep some time for Q&A as part of the session. We want to make sure that we get a chance to hear from all of you because that's how we have been building this product. So before we get to the demos, I do want to take a moment to talk about how we are building before we get to what we have built. So this is-- A lot of this is primarily grounded in listening to our users, not in a planning meeting, not in a roadmap review meeting, but we have, over the course of time, built numerous channels for us to make sure that we are grounded in the community sentiments. We have an advisory board where we meet users internal, external every month, not just to validate the problem, but to also validate the solution and essentially co-build with the communities. We spend a lot of time doing user research. We have ERGs, Employee Resource Groups, within Microsoft, which do a tremendous job in shaping not just the workspace, but also the products that we are building. Lastly, we have a lot of partnerships that we have, we take pride in with BITS and with DAISY, who are giving us tremendously incredible feedback on how to go about building the solutions, right? We have Enterprise Disability Answer Desk and Disability Answer Desk, and both of which gives us the voice of the customers by having the channels that are inclusive out of the box and making sure that the feedback that you submit on those channels reach us, we have Windows feedback, as well as we look for all opportunities to be in the middle of users, right, be it attending events like NFP or World Blind Union or Ignite and Ability Summit that's coming up, or even sessions like this. So we are always looking on ways that we can spend more time with the customers or with the users of our products and make sure that what we are building aligns with the true needs of the community. So that's a quick recap of how we are doing. Everything that you are about to hear, see in the rest of the session comes from these channels that we have built and is informed from the community themselves. So with that, I will hand it over to Nidhi from our team to walk through some of the good work that the team has been doing in the last couple of months. So with that, over to you, Nidhi. Nidhi, we are not able to hear you in case you're speaking. Nidhi, you're still on mute. Can you check? So while we are working on resolving some technical difficulty, feel free to post your questions. We are excited to hear from all of you. on what are the things that in case you have tried Narrator or any top of mind sentiments you have towards any recent experience you have with Narrator. We also have a link that's floating at the bottom that is aka.ms/AME/Narrator, which is where you can post your questions as well.
Speaker 2
Can you all hear me, Ankur?
Speaker 1
Yes, Nidhi, we can hear you now.
Speaker 2
I don't know what happened. Okay.
Speaker 1
All right.
Speaker 2
Thank you so much, Ankur. Hi, everybody. I'm Nidhi. I'm a product manager on Ankur's team, and I'll be your speaker for the next 20 minutes. I am a screen reader user myself. And today we thought, how about sharing five quick reasons why you should try Narrator today, right? Sounds exciting. So let me give you all a quick refresher on Narrator just in case you don't know what it is, right? So Narrator is an in-built screen reader in Windows that helps people with blindness or low vision be more productive, right? Like that is the way we all use our Windows PC. So today, out of the five quick reasons to try Narrator, let me start with the first one. First one is about delightful voices. So it's a simple scenario, right? We all at work, at home and at play, read a lot of documents, emails and articles. And many a time we are, you know, the quality of the voice and the voice itself plays a very important role in how we perceive and understand the content. Right. So here I have a document open and I have turned on Narrator using Control Windows Enter. That's the shortcut in case people are wondering how to use Narrator. And I'm going to read the same content in my document twice, one with the traditional screen reader voices. So let's hear that.
Speaker 3
Modern AI voice generators use neural text-to-speech, NTTS, systems powered by deep learning. These models are trained on massive datasets of human speech and learn to replicate pitch and tone, rhythm, and pacing.
Speaker 2
And now let's hear it in a new voice.
Speaker 3
Modern AI voice generators use neural text-to-speech and TTF on massive data rhythm and pacing, natural pauses, and emphasis.
Speaker 2
I hope you all could appreciate the difference, right? This is what we are calling as natural and natural HD voices that we are offering as part of Narrator. You could choose, of course, between the traditional screen reader voice and the natural voice because the natural voice understands and adapts itself, its tone, its spacing, its emphasis, depending on the text you are reading. So sometimes when you're actually reading content to deeply understand it, you could switch over to using the natural voices instead of the traditional screen reader voices, which are too fast, which are also important. But you have options for you to choose them. These are what we call as delightful natural voices. For you to use these voices, you need to download them for the first time on your PC. And once you have it set up, it continues to work offline. So there's no data that's going to the cloud. We have these voices in multiple languages and not just English. So you could choose the voice and the language that you prefer. So you could preview. There are different voices like Ava, Andrew, Jenny, Aria, et cetera. that you could choose from, and of course, choose a language of your choice. Awesome. So now that we have the voice piece sorted, so you have a smoother experience of how you are actually listening to your content, now let's talk about actually doing some work, which gets me to the second reason of trying Narrator, which is about reading and writing with confidence. As Ankur was mentioning earlier, we listen to a lot of listening channels, and we talk to our customers. We have people who are screen reader users within our company and the team that we interact with as well. A lot of you all said that using Narrator to simply read or write a document is not possible today. What we have done is we have truly understood your feedback to the minute degree, and we've made it really easy for you to create, read and of course collaborate on these documents in the modern day we are all collaborating right so what we'll do now is play a video where it is a compilation of multiple things we've done within helping you to make reading and writing easy and then I'll get to it after that so let's play the video
Speaker 4
Today, a simple task like reading and creating a document isn't doable with Narrator. In partnership with the Word team, we've made it intuitive and efficient to use Narrator with Word. Let's look at improvements in three key areas. First, reading and editing documents is now much more efficient and reliable. Narrator offers continuous reading with smoother, more natural voice feedback, making it easier to follow along. Narrator Word now confirms whether key actions were successfully applied. For example, when I select transportation and press Control B, Narrator announces bold on.
Speaker 5
Transportation selected. B, bold on.
Speaker 3
6.5 selected.
Speaker 5
I, italic on.
Speaker 4
This confirmation works for over 60 scenarios, giving users confidence that commands were applied and reduce uncertainty. Second, Narrator provides clear hierarchical information for tables and lists. For tables, Narrator now gives more precise feedback about your position and the content, with improved awareness of complex or non-uniform table structures.
Speaker 5
Transportation, row 11 of 12, column 1 of 3, merge cell spans two columns. 6.5, column 3 of 3. 1.4, row 12 of 12. Last cell of the table. Pressing tab here adds an extra row. Exit table.
Speaker 4
As you navigate lists, you'll hear the list level and type, making it easier to understand structure and edit your lists accurately.
Speaker 5
Enter list. Heading 1-1. Mitigation and adaptation strategies. Level 2, heading 2-1.1 mitigation. Exit list. Mitigation involves reducing the flow of greenhouse gases into the atmosphere.
Speaker 4
Third, collaborating in Word documents often involves using comments to share feedback and ideas. For sighted users, this process is seamless. A quick glance shows the highlighted text and the comment associated with it. But for narrator users, reading and resolving comments used to be inefficient, requiring a lot of navigation before hearing the actual content and making it challenging to associate comments with their text. Now, we've optimized the experience so you can skip unnecessary details and directly hear the comment, saving time and keeping your flow intact.
Speaker 5
Comment, please attach the link to the video for the section. By Anita Isala on October 22nd, 2024, 02 AM, has no replies.
Speaker 4
With 70% reduction in keystrokes, you can quickly access comment content, resolve tasks, and reply, all without losing context or momentum. With these improvements, Narrator and Word now work together to make document creation, editing, and collaboration more accessible than ever.
Speaker 2
Great, great. We have heard about how Narrator can help you read, edit and collaborate in Word documents and just not just limiting this reading and writing efficiency to the Word documents. We have also added a bunch of shortcuts so that it's quick and easy for you to get to the content that matters for you and not having to read things line by line. For example, We have added commands to read status bar. I think one of the things we've heard loud and clear for the community was we need a command for status bar to read information that different M365 applications put on the status bar. Like if you select some text in an Excel cells, it will give you the sum and average, it will give you the count of words, et cetera. So we have added a new command, narrator key plus backspace, narrator key plus backslash, to help you read the contents of status bar in applications that have one. Similarly, you know that narrator scan mode works not only on the web, but it works across the desktop on any application you are in. You can turn it on and off by just using the narrator key, which would be caps lock or insert in your case, plus space bar. So narrative key plus space bar turns on scan mode, and now you can quickly jump to contents. So for example, if you want to get to a list of steps in your document, which talks about instructions to install something, you could just press L to go to the listed content, use I to just skip to the list item content within it, read it bullet by bullet. Don't worry about any extra spaces, other formatting between it, et cetera. Similarly, if you're reading long tables or long lists, sometimes all you want to know is when does this end or when does this begin, et cetera. So you could use the shortcuts comma and period to just get to the beginning or to the end of the long form table or a list or a land list of landmarks that you have. And I'm sure that all of us come across web pages, which have lot of advertisements and lot of links between the content these days. So you can simply use N to get to the next text content and skip all the extra links and advertisements that come in the middle of your web pages. So together, a lot of compilation of small improvements we've done in Word, which compile together to give you an amazing reading and writing experience in Word documents. And with a lot of new quick navigation shortcuts for scan mode, you can now author and write and read with confidence without having to second guess anything when you are working. Now, now that you have all the text pieces sorted, what about visuals? That gets me to the third reason. The third reason for you to try Narrator is how Narrator helps you understand visuals using AI. So now here I have the same document that I have been using for this demo, and I come across an image. Now, this is an infographic. One of the options for me is to skip this altogether or reach out to a sighted counterpart and ask about the contents of the image. Instead, what I could do is use a new feature within Narrator where Copilot helps you describe the contents of this image. So my focus is on this image, and I'm pressing narrator control D, which is the new shortcut for getting descriptions. And let's see what happens. The image basically gets attached to the Copilot application on your PC, and now you can ask Copilot any questions, ask it to describe it, ask it some follow-ups, et cetera. So let's get through the first part of the video. So my focus is on the image right now. It's an infographic. I was reading a report about a job course, like job research. So let me hit the Control D and let's see what happens.
Speaker 3
Pane, Ganoff, M365 Copilot, quick view pane. Message M365 Copilot, edit. Describe this image. Microsoft 365 Copilot document, message Copilot, edit, space, loading, space, generating response. The image is an infographic from the World Economic Forum's Future of Jobs Report 2025. It visualizes what the global workforce would look like if it were represented by 100 people, focusing on upskilling needs by 2030. Here's the breakdown. 41.
Speaker 2
Awesome. I have just hushed this description for a minute. So what I just did is I hit Ctrl D, the M365 Copilot, since I am using a... Work PC right now, it opened the image that I was focused on seamlessly, got attached to M365 Copilot. My focus was on the text box. I just typed in there description, hit enter, and I got the description read out loud to me by Narrator. Of course, now let me see. Let me ask it another question. Let's go ahead and play around with this a little bit more.
Speaker 3
Can you give me a similar breakup for IT workforce? Loading, space, generating response. Here's a similar breakdown for the IT workforce. Based on insights from the DevOps Institute upskilling IT report and other industry sources, dash of the IT workforce were 100 people 34 would not need major upskilling because they already possess current technical and processing.
Speaker 2
Awesome. So I had a question about how would IT industry do and looks like it has answered those pieces of the questions as well. So great. So I can continue and follow up with Copilot long as I want with more questions about the image. While I'm talking about image a lot here, I want you all to understand that this feature can be used for multiple other scenarios. Say I'm on a website, I am on an unlabeled button or a graphic, and I quickly want to know what could be a potential label of it. I use the same control D shortcut. I it again opens the pilot. I can ask whatever I want and essentially get the information. So I don't need to rely on, you know, ensuring that everybody is doing their part in building accessible websites. Similarly, sometimes I just want to know how's my entire screen? What's the layout of this webpage that I have opened? What's, you know, how does my screen look like? Does it look presentable? Like when I I'm sharing my screen on Teams, like I just want to make sure that none of the... to control S, S for screen, describe whatever you're focused on. So it would again take a screen grab of the entire screen. attaches the prompt that you want to directly have a question about, hey, is there any thing hiding in my screen? You could just directly ask that and then hit enter. Note that we don't directly share any of your private data with the cloud. Like with Copilot, you are in control. You have to explicitly choose send to send the image to Copilot. We are just attaching and keeping it ready for you. While we are doing all of this on all Windows 11 PCs, We are taking it a step further on Copilot PCs. You all know Copilot PCs are this new class of hardware where we have the power of local models, which you can get to your benefit. So the same flow, the same shortcuts on a Copilot PC, if you have one, you use the same shortcut, we use the local model to give you the description, and then only if you have follow-up questions, you can ask Copilot, you see a button called Ask Copilot when that experience opens and you can continue the conversation. So this helps you in cases where you do not want even the first cut of information to be, you know, gone to the cloud. So you could use your Copilot PC for that. So overall, text was sorted. I now also know that I don't need to skip visuals, or I don't need to depend on sighted colleagues for this work. So it's definitely getting me more productive so far. Let's see what more I can share, which brings me to the fourth reason, adapting narrator to what I want. Now, if you are a screen reader user, you know that verbosity could come in your way, right? Like, sometimes your screen reader could feel chatty. What I'm doing right now is I have an Excel open here, and it's reading a lot of content for me. This is a capacity plan that my engineering manager, Sri Hari, who is also on the call, has sent me, and I'm reviewing it. So my job here is to quickly read and review. So let me start, Narrator, and see what I hear.
Speaker 6
Team capacity Q1, Excel window.
Speaker 2
Awesome. The Excel opened. It did not read what I wanted to, but on a normal day, it would have read a lot of content. And what I really wanted to hear was not a lot of content. I just want to hear the name and the value of the cell. So now we have introduced a new feature called Personalization in Narrator. So you just quickly-- whenever your focus is on an element that you want to control its announcement for, you keep your focus there. Give the command to Alt+B. That opens this new personalization window. It opens with a text box. I can just ask what I want to be personalized for me. So here, I'm going to just ask announce name and value instead of the entire information from the cell.
Speaker 6
Customize narrator window on Microsoft. Announce. N. E. E. Y. Name.
Speaker 2
Send button.
Speaker 6
Analyzing instruction. Analyzing instruction. Analyzing instruction. Instruction understood. Preview. Product manager. B2. Save and close button.
Speaker 2
Awesome. So I heard that it has understood my instruction and it said product manager B2, which is like a very clear announcement. So I've saved and closed my window. Now let me go back to my Excel and see the new announcements.
Speaker 6
Save and close button. Team capacity Q1, Excel.
Speaker 2
Great. You rehear it very clear. I think the audio has some issues yesterday. So maybe let's see if we could clean it up for the second scenario here. So what I did in hearing a lot of content in the cell, I just wanted to hear what's the name and the value of the cell and not things like the coordinates in the Excel, the table headers, et cetera. So I just said and asked that editor in the new narrative personalization feature to just announce name and value, and it just did that. Now, today, let's think about another example. Here, I'm not just looking at scanning the Excel, but I really need to validate the data that somebody has sent me. So it's a very different scenario. I need to be hearing a little bit more content. So generally, what I'm hearing is this on this sheet now. So it's a new sheet, new Excel, so let's see what I hear. Great. So it's telling me some details, but I did. OK, so it's in minus 5000 B6. It's continuing to read the name and the value, which is great. Although I'm not sure if where this minus 5000 is coming from. Is there a formula my team is using to put this? data. So let me again bring my narrator personalization window and ask there to announce if the cell contains a formula.
Speaker 6
Customize narrator.
Speaker 2
Announce value.
Speaker 6
Analyzing instruction. Instruction understood. Preview 20,000 contains formula. Save and close button. Business report Q1 2020 minus 5,000 contains formula. 20,000 contains formula.
Speaker 2
Great. So I heard that these minus 5,000 minus 20,000 are details of a formula. I could click F2 here directly to get into what the formula was and get into my validation mode. So the same feature allowing you to customize your experience depending on what you want to hear, depending on the task and the application you're using. One thing I do want to note is, although I was showing you the way to quickly do this only using natural language typing in a text box, we do have below the new UI that opens for personalization also check boxes so that you could manually go and enter and check and uncheck different components of announcements that are happening. So for example, your values, your cell numbers, if you want something to be announced as unchecked at the beginning instead of later. So you could also swap the order in which some of these announcements come, depending on your liking. So all of that is about adapting Narrator to your needs. I am definitely feeling great today. I have beautiful voices that I'm using. I am feeling confident in the documents I'm authoring and reading. I do not need to depend on sighted people for visuals, and I'm also personalizing Narrator to what I want. Let's see what more we have added in Narrator that will help your everyday tasks be more efficient, and for that, I will hand it over to Roshni from my team. to get us on that.
Speaker 7
Hi, everyone. Is this, have I unmuted now?
Speaker 2
Yes.
Speaker 7
Okay, thanks. So hi, everyone. My name is Roshni, and I'm going to take you all through the fifth reason to try Narrator, which is the everyday essentials that we might need in our day-to-day scenarios. This is to stay in control, stay efficient, and very specific to different needs that come up while using screen readers. So first, I'd like to start with a scenario. It's a situation where we want to hear narrator again, repeat what narrator said. There's cases where we're dealing with unfamiliar text, new names, weird, clunky error codes, and it goes by without like, and sometimes it's hard to, really parse what has been said so in in these cases there's um you we can hear the last phrase which is uh rehear the last phrase or even copy the last phrase depending on our use case sometimes it's useful to copy when we need to save the details for later like suppose it's someone's address for an event or it's a meeting link or it's actually an error code which we need to send to support so It's often helpful to rehear our phrases if we have difficulty parsing them or copy them if we need to save them. And the shortcuts for these things are narrator X to rehear it and narrator control X to copy it to your clipboard and then you can paste it wherever we like. The next thing is that sometimes just reading the last phrase isn't enough. We would like, suppose there's situations when we're really completing a task and then somebody's talking to us, And we've lost track of what narrator said, or we just got distracted and we lost a couple lines of what narrator said. In these cases, speech recap is pretty helpful because it's a window of the last 500 words spoken by narrator, so a window pops up and we can navigate in that window. through the log of what narrator has recently spoken to quickly place ourselves back in our flow. So the shortcut for narrator's speech recap is narrator alt X, and that brings it up, and it's a pop-up window. And what's... What I've heard is that a lot of times sighted teachers also use the speech recap and they clip it to the right side of their screen so that when the students are using headphones to privately listen to the screen readers, then it's often helpful to like follow along visually about what narrator is saying so that we don't have to play the audio out loud on the screen. So that's another helpful way to go about it and It's, yeah. And the next thing is that if we're, if we have, for users who are using a refreshable Braille display, we have a similar Braille viewer which displays, which is a pop-up which displays what is the Braille and the text content currently being spoken by narrator live. And this is also especially helpful for instructors, sighted parents, and anybody who's supporting the learning journey for a user who is using a refreshable braille display to keep context of what's going on. So next, so these were two things. One was to, so this is different ways of accessing narrator's speech output is that we maybe we want to rehear, maybe we want to copy it, maybe we want to have a bigger log of what was just said, maybe we want to, see the visual representation of what the Braille, while Braille is being read as output from your screen reader. So the next thing is I'd like you to walk you through two features that are designed around privacy and focus. The first one is screen curtain, which is a pretty popular feature request, very widely requested, and it's when, and we can use this feature to black out our screen. while keeping narrator running. So this might, this is useful when we're in public places like cafes and we don't want anybody to be shoulder surfing. Or we could just, I think it's also helpful to just save battery to not have your display on all the time. So to toggle on and off narrator, sorry, not narrator, but screen curtain, you can use narrator control C and that's to toggle screen curtain on and off, which would, either make your, it would darken your screen or it would revert that change. While you're using screen curtain, you can share your screen during meetings. You could take screenshots. All of that is none of the functional functionality is blocked, but it's just a physical overlay on your screen. The next thing that I'd like to talk about is speech off, which is a, which is a recent, development and pretty exciting is and it's been a scenario where we have multiple audio streams, right? Where we're in a meeting and a lot of people are talking and or we're in an application where a video is playing or an audio is playing and we don't want a screen reader to overlap with that, but we want this screen reader to remain active still. So these are cases where we temporarily shut down narrator's voice while narrator is active and We can choose to just engage with the audio or the video or like in the other voice that's being part of our environment. So for speech off, the shortcut is narrator shift S. This is both to toggle it on and toggle it off. So we're coming up with pretty cool, new, exciting features. Some of it, a lot of it is to fix our fundamentals, a lot of it is to improve it, a lot of it is using the cool new, all the AI features. And we have recently added a way to inform, like, for users to stay updated on what's the new things, what are the new things that we're releasing. So that's part of our what's new experience. which is a dialogue that opens when you turn on narrator. And this dialogue is a series of pages that walks you through different narrator updates, performance improvements, feature updates. And also, you can reopen this dialogue anytime from settings, from narrator settings, and take a peek at what are we up to and what are we releasing. I'm super excited about for the next few months where we're gonna be releasing a bunch of new things and I'd really like to get your feedback on those. And so that was, that was the, that was my part of the, which was everyday essentials, which was, can we access spoken output again? Can we, can we, be more private with our screen curtain if we want to. And if we have multiple audio streams, then we can maybe turn off our speech and focus on the audio at hand. So and the last thing was the narrator updates, which helps us understand what's the new and fun features that we've just released. So now I'm going to hand it back to Nidhi to to take it forward. Yeah, thanks.
Speaker 1
Nidhi, you're still on mute in case you're speaking. Just give another second to see if Niti comes off mute. All right, so maybe I'll take it while we wait for tech issues with Niti to be resolved. So hopefully, last 20-30 minutes gave a glimpse of the spectrum of improvements that we have been working on later. And this is a lot of hard work that team has been putting in. And hopefully folks here appreciate the level of details that many of these features need us to get into. So whether it's, you know, the sounds of how Narrator comes across reading or writing with confidence across word and different ecosystem, or be it understanding visuals with AI, whether it can be local AI or it can be Copilot, or how Narrator is starting to adapt to your needs so that you have ability to influence the level of verposity and make sure that it's customized to improve your productivity at the end of it. And then lastly, everyday essentials, whether you're in a public place, screen code and all the good stuff that Rosni has talked about, right? So again, this is still not a comprehensive list of everything that the team has done, but we thought we'll just come in here and talk about some of the big wins that we believe are going to be meaningfully improving the experience of our screen reader users. So with that, I think we'll probably just talk about some of the resources. What we have on the screen are two QR codes. The first link is for you to stay updated with the latest improvements and to share your feedback with us. This is aka.ms/coding. Contact Windows Accessibility. Again, I'll repeat, this is aka.ms/contactwindowsaccessibility. This will open up a form that allows you to share your feedback regarding anything on Windows accessibilities, right? While we stood here today talking about narrator and screen reader ecosystem, we welcome any feedback that you have on the broader Windows accessibility landscape. The second QR code that I have here is to This is to help you learn about all the resources, right? If you're curious on learning more about the accessibility resources or the improvement that we are talking about, this is the second URL, which is aka.ms/windowsaccessibilityresources. Again, I'll repeat, aka.ms/windowsaccessibilityresources. This is where you'll find comprehensively all the new improvements that we have been working upon. So if you are looking to learn about anything on the Windows accessibility landscape, we do have a lot of good resources that allow you to learn about the current features. A lot of things that we didn't show today also is on that link, as well as a lot of new work that we have been doing is all captured in here. So that's a place where you can start for learning about the accessibility resources. Awesome. And then I think I'll just repeat the link for submitting your questions. We see some questions coming in already, but just for the sake of comprehensiveness, I'll repeat that. It's aka.ms/ama/narrator. This is where you can go and submit your questions, and we'll make sure that we carve out time. In fact, for next 20 minutes or so, we want to make sure that we spend time with all of you answering some of the questions that you may have top of mind. Awesome. Am I missing anything, Rashni? Anything else? Can you jump into the Q&A now?
Speaker 7
I think Edad would be a good mention.
Speaker 1
Edad, essentially, I think I mentioned it briefly. So we do have a disability answer desk and we have an enterprise disability answer desk. Those are. dedicated channels for us to listen to our customer feedback, right? And this is the channel that we very, we make sure that we pay attention to any of the feedback coming in from those channels. So if you have any feedback, it's also the workflow is extremely optimized to make sure that you don't feel any friction in order in the process of submitting feedback. So again, right, I'm ensuring a lot of resources, but that's the reality, right? We are trying to open up as many channels as we can, so that we have ability to understand the sentiments, understand the top of mind questions, concerns that you all have. And so that we are able to make sure that the product evolution is in alignment with the customer, with the, sorry, with the community sentiments. And yes, there are multiple, EDAR gives you chat, phone number, sign language. In fact, we have partnership with Be My Eyes app as well. that you can leverage as part of giving the feedback. Awesome. So I think with that, we will open up for questions. Our agreement was for Nidhi to answer all the tough questions. So we'll give our best shot. I think Nidhi is also coming back into the call. So let's see, I think we have some questions that were getting queued up.
Speaker 2
Can you hear me?
Speaker 1
Yes, Nidhi, we can hear you.
Speaker 2
Okay, awesome.
Speaker 1
Thank you for covering me. Yeah, no worries, Nidhi. I think we just covered the rep content. We are just jumping into the Q&A bit. So you have joined at the right moment. And Roshan, do you want to start help with the moderation of the questions coming in? Or do you want me to take that?
Speaker 7
Sure. I can do that. The first question is, can you provide updates on Braille support for Narrator? In the past, users have had trouble connecting Braille displays due to driver compatibility issues. Has this been resolved? I would like to use my Braille displays with Narrator and other screen readers and connect and disconnect between different screen readers on the fly. Is this possible?
Speaker 1
Amazing question. Nidhi, would you want to take that?
Speaker 2
Yes. So first of all, we want to acknowledge that we've done some work for the driver conflicts. We've heard this feedback a couple of times from a lot of users. And so a few years ago, actually, we shipped a filter driver solution that helps you actually use your Braille displays, switch between screen readers. But I also understand that it is that solution is not working for certain kinds of devices. So if you have like a composite Mantis Q40 or a HIMSS display or a Humanware display, there are a few classes of these devices with which we are hearing that the driver conflicts continue to exist. What I could say today is we have something exciting coming really soon for all of you without sharing a lot about what it exactly is. But maybe by the time we are here for the next webinar and please watch out for this announcements pretty soon at Ability Summit as well happening at May. So with that, I would say we acknowledge and we are actively working on different solutions to close out these driver conflicts so that you can seamlessly use your screen reader and your Braille display of choice with Narrator.
Speaker 7
So, Rashin, you want to go to the next one? Yeah, the next one is... more feedback than question, but so it's feedback from the tech community page, which is I'd like to see automatic language switch in Microsoft Narrator so it can read multiple languages at the same time, like English, India and Hindi language at the same time.
Speaker 2
Yeah, I'll go with that one as well. So today we support automatic language switching as long as There are two things that are satisfied, right? One, the author has ensured the language tag. So for example, if it is English, India, or Hindi is mentioned in the content, and you have the appropriate voices installed on your PC. It could be any of the voices, right? Your natural voice or your any other voice of the screen reader that you're using. What we are hearing of late is a lot of document authors are actually not doing the right thing of tagging their content. So we acknowledge that feedback and we put that and are looking to prioritize that as well in our backlog. So, yeah. But thank you so much for trying to, you know, try and use the reader with different languages. I'm sure for most of the documents and web pages out there, it'll work for you, but in case there are specific web pages, and documents, we would also like some feedback from you all. So if you could either use the Disability Answer Desk that Ankur spoke a little bit about earlier, or just use the Windows Feedback Hub with Night Alt F or Windows F. It is Windows F key for Feedback Hub, and you could choose Narrator. And if you could drop certain examples of documents or web pages that are not working for you, that will help for us as well. But as I said, as long as the language tag and you have a voice installed with that language, we do switch and ensure that you get a seamless experience of hearing that language.
Speaker 7
Thank you, Nidhi. The next question is, From LinkedIn is, isn't it a step back to the right? Isn't it a step back to write commands with text to control the verbosity? Isn't it better to use shortcuts for that or dialogue boxes?
Speaker 2
Yeah, understood. I'll take that as well, so we... We have multiple things that we are offering here, right? So for example, you still have the verbosity shortcuts available for you. So your narrator key, we cycles through the different verbosity levels that you have with Narrator. We have five verbosity levels, one to five. Your default is three. So we cycle between all of that, right? What the dialog box tells you is do a, go a extra level, right? So for example, I can configure as a product manager, come up with smart defaults and say one to five means this, this, this, like configure this, don't read this, don't read that. But all of us, as I showed in the examples that I was using, depending on the task at hand, I sometimes wanted to hear more. I sometimes wanted to hear less. And so we made it easy for you to say, while you have the default verbosity levels, you could customize it a step further, use the box. The dialog box is a way to get it done because a lot of us are now using natural language to express ourselves better. So you could just use that. But you also, below that, get a lot of second-level details in terms of all the fields that are getting announced for that control type. So you can go check, uncheck, like really go detailed in it, right? So yeah, to answer your question, we do have shortcuts. If you don't want to get your hands dirty and go deep, use the Nerator V verbosity settings shortcut, or if you want to go deep, use the Nerator Alt P to personalize your announcements further.
Speaker 1
Yeah, I think you covered it well. I just want to add that I think the video when it played, it was a little bit choppy. So maybe that's where some people possibly missed it. But as you rightly pointed out, Niti, that there is both. You don't need to just go via natural language, but you can actually just manually. go through a couple of checkboxes and have a stronger and a cleaner control on what to announce, as well as hear the preview of what it would change it to, right? So we have certainly got this feedback early on and did course correct on the way we've built the solution. Okay, Rashmi, next one.
Speaker 7
Let's go to the next one. Next one is feedback. We'd like the ability to use other synthesizers with Narrator, any work being done on this area?
Speaker 2
Yeah. So as you notice, we are actually using multiple synthesizers today. So we have the entire one core set of voices that we call, which are the Microsoft voices, David, et cetera, et cetera, which comes with that set. What we showed you today were called natural HD voices we are calling, and then there's a natural set of voices. So today we do support multiple synthesizers. We also understand that users are looking at even other synthesizers that are available and supported in other screen readers beyond the Microsoft and Natural Voices we've spoken about today. We've heard this feedback, we are acknowledging it, and yeah, we'll share in case we have an update on that.
Speaker 7
Thank you. The next question from on LinkedIn is, have you noticed any differences between using the desktop versions of Word and the other Office apps and the online versions? Has the experience improved with the web versions?
Speaker 2
Yeah, so general recommendation that we do is if you are a screen reader user, we prefer, or in fact, even Office preference, we all use the desktop versions. Having said that, there's a lot of improvements that are happening on the web apps as well for the M365 applications. You should all check out the support documentations for that. And there is definitely a lot of improvement coming along there as well.
Speaker 7
Thank you. The next one is, do HD voices work offline, or do I need an internet connection?
Speaker 2
Yeah. So they work offline. You just need to set it up for the first time. So the first time you go to narrative settings and download them, you need an internet connection. But once you have it set up, it all works locally on your PC, Windows 11, Copilot PC, any PC that you have.
Speaker 7
Got it. The next question is, I'm doing some, I don't know what THON, THON based accessibility audit to test the quality of the alt text. Can I use the Copilot with narrated image description in an automated way?
Speaker 2
Got it. What you could do is You could use a different feature, which is not part of Narrator, but your Office application in case you have these images in Office applications. And if you have a Copilot PC, which is a PC which supports local models, that gives you auto alt text basis Copilot itself. So that would help you get auto alt text as soon as you put a bunch of images in the document. So that is a way you could work around the situation. I mean, not necessarily work around, but get the desired result here. But what we have currently with Narrator and Copilot is you have to, on-demand, keep focus on whichever image or control that you want to get described and do it one by one. But that's a pretty interesting use case that you have shared, which we could take back and think a little bit more about as well.
Speaker 7
Alright, let's go to the next one. I find narrator interesting, but it is a webpage like structure and in a webpage format it can be hard to find options. Do you have tips? I also wanted to ask how to open personalization setting in narrator.
Speaker 2
Got it. The personalization, the video we showed, it's narrator key, Alt P. So caps, Alt P, mostly you would have your narrator key set to caps or insert, which is our default setting. Now, when it comes to web pages, there's a lot of tricks, right? The first thing is, of course, scan mode, which is on by default when you are using narrator on a web page. use quick navigation keys. So for example, the first thing people look for is a skip content link, which the web pages provides at the typically at the top. So you do control home and you reach at the top and you get that get to that skip content link. Once you get that out of the way, people do a little bit of, hey, what could be the summary of the web page? A little bit like just to understand the layout, et cetera. So there are two things you could do. One, you could just typically rely on heading structure. So your H command or your one, two, three commands, depending on if you want to just go to the next in the previous heading or next in the previous heading at a certain level. So one, two, three would take you to the headings at level one, level two, level three, or H will take you to just like all headings on the page, right? Now you could do that. So that's the quick shortcuts. The second way is you just bring up a list of list of links and list of headings, dialogue. So this is narrator key F5, narrator key F6, that will give you list of different elements on the screen, right? So F5 gives you landmarks today, F6 gives you headings, F7 gives you links. So that's narrator F5, F6, F7. Although these are like traditional ways that screen readers are, screen reader users are taught to get a fair, quick understanding, scan, understand the web page, and then get to the content, right? So we But I would share a different perspective as well. These days, people do not want to navigate and they still quickly want to use summaries. So you could use the Copilot descriptions that we've showed. Also, if you're using Edge, you could use the Copilot, which comes in Edge itself. That helps you quickly do some Q&A because that has interesting understanding of your entire web page as well. So these are different ways. I've covered a little bit of traditional hints and shortcuts, and then of course you could use the summarize features. And just to remind you, it's Narrator key control S, which is caps control S for your full screen summary as well on Narrator.
Speaker 7
All right, so that's, we're wrapping up Q&A now. And thanks everyone for your questions. You can, we will be looking at both tech community and LinkedIn for at least the coming week and next. For any questions that you might have, we can, we will be, we can, we will answer them back on the chat thread itself. And if you would like, to get in touch with the team, definitely fill in aka.ms/contact-windows-accessibility, aka.ms/contact-windows-accessibility. If you put in your e-mail and give us suggestions, then we'd love to have you part of the broader conversation on what we're doing next and things like that, all of the exciting stuff. I'm going to hand it back to Ankur for a minute to wrap it up for us.
Speaker 1
No, this was fantastic. Thank you. Thank you, everyone. Thank you for joining the session today. We are very excited to have you along with us on the journey, and we look forward to sharing more improvements ahead. If you are tuning in on demand, we leave the Q&A open so that we have this channel open. But I'm hoping that some of the resources that we share gives us a longer connection in the long run so that you're able to share your feedback. If you have any feedback on this session on how to do this better and how to make it more useful, please send them our way. And we are in this journey building this together with all of you. And with that, I'll say thank you everyone for being here. Thank you for all the feedback that you have given. Thank you for all the questions and we look forward to continuing this journey ahead with you. And with that, we'll sign off for the session today. Thank you.