Welcome to the sixty-eighth episode of Access On, the National Federation of the Blind's Technology podcast.
Episode
Listen to the sixty-eighth episode of the Access On podcast (Browser).
Or listen on your preferred podcast platform.
Timestamps
This week on Access On:
- Looking ahead to our Apple 50th anniversary special 0:00
- Register for our visual interpretation webinar 2:22
- Robin Christopherson demonstrates and reviews Omni Describer 3:16
- Brava Smart Oven 20:39
- Connecting my Windows PC to an Android TV 22:33
- How accessible is Preply? 25:58
- Can Graphiti act as a computer monitor? 30:32
- Waymo ride experiences 33:21
- Accessibility frustrations with the Facebook website 35:17. To provide feedback to Meta, please use their accessibility feedback form at: https://www.meta.com/help/policies/707685072208748/
- Accessibility of delivery robots 40:46
- Unwanted verbosity in Zoom for Windows 42:10
- Keeping my iPhone charged to 80% 44:44
- WhatsApp for Windows accessibility bug 47:38
- Website administration 49:27
- Accessibility of Webex meetings 51:18
- AI agents in screen readers 52:09
- Tech Tip: Using JAWS speech history to overcome accessibility issues 56:12
- Closing and contact info 57:33
Transcript
Jonathan Mosen:
Welcome to Access On, the technology podcast of the National Federation of the Blind. On this week's episode, customized machine-driven audio description with Omni Describer. There's plenty of listener feedback, including how to use Miracast to send something to a big screen.
This mounting frustration about Facebook accessibility, when will we see sustained progress? And how much power should a new generation of screen reader AI agents actually have?
It's Jonathan Mosen at the Jernigan Institute in Baltimore, Maryland, welcoming you to episode 68 of Access On. A show note, a heads up as they like to say. First of all, typically Access On episodes come out like clockwork.
Like clockwork, I tell you at 2 o'clock every Wednesday, Eastern time. Next week will be no exception. And Wednesday of next week will be April 1st. And no, we're not going to fool you on Access On, I can promise you that. We're going to have all legitimate things happening on that day.
And the episode is going to be devoted to something that we're celebrating on April 1st, and that is Apple's 50th birthday. The story of Apple's second wave of accessibility is well known and well understood, and it's been told many times.
And it starts with voiceover coming to the Mac first of all, and eventually to iPhone. And of course we're still living that now. But there was another golden age of Apple accessibility that was very much blindness driven in the 1980s, a period before some of our listeners were even born, but some of us were using technology back then. Oh my goodness, I'm getting ancient.
So we are going to be having a look at this next week. And if you remember things like Braille Edit and Becks and the Echo II and the Cricket Synthesizer and various other things of that kind, then you will be reminiscing with us next week. And if you haven't heard of any of this before, we have a tale to tell. So I'm looking forward to bringing that to you next week. We've got some great guests for that particular feature.
And the day before that, so it's pretty close now, on March 31st, we have another Access On webinar. This one is happening from 1:00 until 2:30 PM Eastern Time on March 31st, and it's all about the wonderful world of visual interpretation. There's plenty of interesting things happening in this space right now. So the National Federation of the Blind Center of Excellence and Non-Visual Accessibility is pleased to give you an overview of what's going on.
If you would like to register for this Access On webinar, head on over to nfb.org/cena. That's nfb.org/cena. Choose the events and training link, and you will find the link to register for that webinar. If you can't attend, but you'd like the full recording, then register and you will receive a full recording of the Zoom event. We will also eventually get to serializing it on Access On here and giving you some highlights.
Robin Christopherson:
Hey guys, Robin here. I'm going to be demoing the amazing utility called Omni Describer. That's two words. And it's free. And it helps describe add audio descriptions to videos that don't have them. And you can do this on YouTube videos or you can do it for local videos that you've recorded yourself and it's brilliant.
So let's have a quick look. I'm in the interface of the app. I'll tell you how you can download it at the end. And I've slowed down my screen reader, so I hope it's slow enough for you.
Speaker 3:
Local video file button.
Robin Christopherson:
Local video file. So that's the first thing you land on when you open up Omni Describer, and that will enable you to browse on your computer to a file that will then be audio described if I tab.
Speaker 3:
Direct video URL button. Alt plus U.
Robin Christopherson:
Direct video URL, that's somewhere on the internet. Not YouTube because that's the next one.
Speaker 3:
YouTube video URL button. Alt plus Y.
Robin Christopherson:
We'll be using that one in a second to describe a very quick fast moving video from Oakley about their new version of the Meta glasses. Tab.
Speaker 3:
Optional: Selected prompt preset to enhance generation: combo box. Left paren, no preset selected, right paren. 1 of 11.
Robin Christopherson:
Okay, this is really important, guys. There's 11 preset prompts, and it's important to choose the right one for the kind of video that you're using. Now, if it's not super fast like this one, you might get away with just none, which is the first one here. Let's arrow down.
Speaker 3:
Movie trailer, 2 of 11.
Robin Christopherson:
That's actually the one I'm going to choose for this video because it's so fast moving. So this is the second of 11. If I tab, it will describe about it.
Speaker 3:
Read only edit. This is a movie trailer. Focus on fast, impactful, and iconic shots. Prioritize describing key action moments, dramatic character, introductions and any on dash screen texts like release dates or studio logos. Create a sense of excitement and hint at the plot without revealing spoilers. Descriptions must be extremely brief to fit between quick cuts.
Robin Christopherson:
Okay. So a lot of thought has gone into each prompt. I'll shift tab again.
Speaker 3:
Optional: Selected prompt preset.
Robin Christopherson:
That was the second one. Let's arrow down and hear some more.
Speaker 3:
Action movie, 3 of 11.
Robin Christopherson:
I could tab to get a big description about this one, but let's keep arrowing down.
Speaker 3:
Romantic movie, 4 of 11.
Robin Christopherson:
This is good, this one. Let's read this one.
Speaker 3:
Read only edit. This is a romantic movie. Prioritize describing the character's proximity to each other. Meaningful touches left paren like holy hand, right paren, and significant facial expressions that show affection, longing, or conflict. Describe the romantic atmosphere of the-
Robin Christopherson:
I'll stop it there. Let's shift tab up. And carry on down.
Speaker 3:
Comedy/sitcom, 5 of 11.
Robin Christopherson:
That's good. It talks about focusing on antics and I don't know, actions in a comedic way.
Speaker 3:
Sci-fi/fantasy.
Robin Christopherson:
Sci-fi, this is a cool one. It says focus on describing machines and spaceships and how they work and what they're doing. That's super cool.
Speaker 3:
Mem/viral video. 7-11.
Robin Christopherson:
Viral video, don’t know.
Speaker 3:
Documentary/educational. 8 of 11.
Robin Christopherson:
You can look at the description of all of these.
Speaker 3:
Cooking/recipe. 9 of 11. DIY/tutorial. 10 of 11.
Robin Christopherson:
Step by step, focus on step-by-step and the before and after effects and stuff.
Speaker 3:
Gaming/let's play. 11 of 11.
Robin Christopherson:
Don't know about that one. Let's go back up to movie trailer.
Speaker 3:
Movie trailer. 2-
Robin Christopherson:
And I'll keep tabbing.
Speaker 3:
Read only. Status logged at read only. Contains text.
Robin Christopherson:
I'm not going to worry about this pain, but this comes up with information in other places. And it's always here, it's like a status pain. And every message that comes up in here is spoken out by the screen reader. Let's tab again.
Speaker 3:
Settings... button. All plus S.
Robin Christopherson:
Let's have a look in the settings.
Speaker 3:
Omni Describer settings dialogue. General tab selected.
Robin Christopherson:
There are three tabs, general.
Speaker 3:
AI settings tab selected.
Robin Christopherson:
AI.
Speaker 3:
Audio output. Tab selected.
Robin Christopherson:
And audio output. Okay, let's arrow back to general. And tab.
Speaker 3:
YouTube video download quality: combo box, 360P, 5 of 7.
Robin Christopherson:
I'm just going to leave that as a default.
Speaker 3:
Application language left requires restart right: combo box. English, 3 of 11.
Robin Christopherson:
What's the UI language that you want?
Speaker 3:
Logging level: combo box. Info left paren.
Robin Christopherson:
Not going to touch that one.
Speaker 3:
Allow descriptions to interrupt current speech, checkbox not checked.
Robin Christopherson:
Okay. So this is important. Do you want your descriptions to overlap dialogue? No, thank you.
Speaker 3:
Okay button.
Robin Christopherson:
Okay, that's the okay tab. Let's go round to the tabs again.
Speaker 3:
Cancel button. General AI settings tab selected.
Robin Christopherson:
I'm selecting AI.
Speaker 3:
Gemini API key: passworded it. ***.
Robin Christopherson:
Okay. Now when you install it, if you look in the help file, then it will tell you how to go and get that key. It's really easy to get it. It will take you straight to the page and it's a code that you just paste in here. Super easy.
Speaker 3:
OpenAI APIkey left paren, TTS, right paren: password edit.
Robin Christopherson:
So I haven't done this one. This gives you ElevenLabs quality, really human sounding speech. So we're going to be using the SAPI 5 engine from what Microsoft offers, but you certainly can do this as well.
Speaker 3:
Frame rate for AI analysis: combo box. No change left for an original FPS right paren. 1 of 4.
Robin Christopherson:
Going to leave that.
Speaker 3:
Description verbosity: combo box. Detailed-rich descriptions. 3 of 3.
Robin Christopherson:
Okay, there are three here, that's the most detailed. If I arrow up.
Speaker 3:
Standard-balanced detail left parent short-concise descriptions. 1 of 3.
Robin Christopherson:
I'm going to put it on the third one again.
Speaker 3:
Detailed-rich description.
Robin Christopherson:
But I think this is overwritten by the prompt that we looked at a moment ago. Tab.
Speaker 3:
Gemini model override: edit.
Robin Christopherson:
Don't know what that is.
Speaker 3:
Temperature: edit. 0.3.
Robin Christopherson:
Don't know what that is.
Speaker 3:
Send video only left paren, no audio right paren, to AI checkbox not checked.
Robin Christopherson:
I don't know why you'd want to do that. Just the video. But anyway.
Speaker 3:
Disabled safety filters left paren. Use with caution right paren, check box checked.
Robin Christopherson:
Don't know what that is.
Speaker 3:
Enable video chunking for long videos checkbox not checked.
Robin Christopherson:
And you'd probably want to do that. Not sure why, but we're not going to use a long video here.
Speaker 3:
Okay button. Cancel apply button.
Robin Christopherson:
And let's go to the third tab.
Speaker 3:
AI settings tab selected. Audio output text-2-speech engine SAPI 5 left paren 64-bit voices right paren, radio button checked. 2 of 4.
Robin Christopherson:
Okay. I'm going to leave it on that at the moment because like I say, I haven't got a key.
Speaker 3:
SAPI 5 left paren 64-bit right paren settings voice: combo box. Microsoft Hazel desktop-English left paren great Britain right paren.
Robin Christopherson:
We could use Hazel.
Speaker 3:
SAPI 5 left paren 64-bit-right-paren settings voice: combo box. Microsoft, zero desktop-English left paren United States right paren. 2 of 2.
Robin Christopherson:
So there's two SAPI 5 choices. Let's leave it on that one.
Speaker 3:
SAPI 5 left paren 64-bit-right paren setting speed: left-right slider. 40%.
Robin Christopherson:
Let's leave that on 40% for the rate.
Speaker 3:
Test voice edit. This is a test of the selected voice.
Robin Christopherson:
Okay, that's the text that's going to be spoken if I turn.
Speaker 3:
Test voice test button.
Robin Christopherson:
Test.
Speaker 3:
Original audio-DB.
Robin Christopherson:
Okay. I think that sounds fine. Tap again.
Speaker 3:
Okay button.
Robin Christopherson:
Okay, that's all the settings, guys.
Speaker 3:
Omni Describer-V1.
Robin Christopherson:
Now, there are a couple of menus, there's a file menu and a help menu. In the file menu, if I do that.
Speaker 3:
File menu. Open local video.
Robin Christopherson:
You've got the same options as you have with the buttons.
Speaker 3:
Open video from URL. Open from YouTube. Manage prompt presets.
Robin Christopherson:
Now this manage prompt presets is the only thing that you can't get to from the main UI in this file menu. And if I hit enter on this, which I won't at the moment, you can add your own custom prompts to that list or you can edit the ones that are there to tweak them even more to your liking.
So if you want to edit those 11 default prompts or add more of your own, then this is where you come. So let's hit escape out of that.
Speaker 3:
Leaving menus, settings....
Robin Christopherson:
Back on the settings button tab.
Speaker 3:
Exit button. Alt plus X.
Robin Christopherson:
And we round the exit button, tab one more time.
Speaker 3:
Local video file button.
Robin Christopherson:
Back up to the top. So let's put in a YouTube video.
Speaker 3:
Direct video. YouTube video. Enter YouTube video URL dialogue. Enter You... Pasted. Https://www.youtube.
Robin Christopherson:
So it's in there. Let's hit enter and it will go away and grab it and start the describing process.
Speaker 3:
URL entered. Downloading from YouTube... Omni Describer -V1. 3, 1 panel. Status logged at read only. Contains text.
Robin Christopherson:
So we're in that edit pane that I mentioned about before and it said it's going and downloading the video. And every time another message-
Speaker 3:
Asking AI to generate descriptions...
Robin Christopherson:
... is put in there, it's automatically echoed.
Speaker 3:
Video is active.
Robin Christopherson:
Okay. So even though it's only a 30 second video this one about the Oakley Meta smart glasses, it will take a while to do its thing.
Speaker 3:
AI is thinking... Successfully parsed 12 raw descriptions, correcting timestamps and duplicates... Timestamp correction, complete 12 descriptions processed. Zero head timestamps adjusted. Finalizing descriptions... Success. Generated 12 audio descriptions. Identified four characters. Video player with descriptions dialogue. Play button.
Robin Christopherson:
Okay. So it's created 12 descriptions in this 30 second video. It's tried to keep them short because this is a fast moving video and it's identified four characters. Now it actually names characters and I've got no idea if that's because it goes away and identifies people from the internet because they're famous people or whether their names pop up.
I don't know, but it often puts names to characters in videos, which is super cool. So now it's dropped us into this player. So we're on the-
Speaker 3:
Play button.
Robin Christopherson:
... Play button. Now here we will hear the MP3 or the audio of the video. And we'll also hear the audio descriptions, but not in the SAPI 5 or the ElevenLabs voice. This is in Ava. This is using Ava because it's a similar panel that pops up the audio descriptions as the panel that we heard a moment ago that was voicing the notifications of progress, et cetera.
When we export it as an MP3, it will burn those in using the SAPI 5 engine. I hope that makes sense. But the reason why we're in this player at the moment is because you can actually adjust the descriptions if you're not happy with them. So let's hit play.
Speaker 3:
Pause. Pause button. Boo Johnson, skateboarder. JR Smith, golfer, I shouldn't wear skates. Bicho Carrera, Paraglider.
Speaker 4:
Play my hooping playlist.
Speaker 3:
JR Smith Dogs. I should wear skates fast. Bicho Carrera paraglides.
Speaker 5:
Hey Meta, take a video.
Speaker 3:
Skater jumps fares.
Speaker 4:
Hey Meta, how strong is the wind today?
Speaker 3:
The wind is blowing.
Robin Christopherson:
I'll just stop it there. So wow, fast action was really good at doing very succinct descriptions. Obviously if we'd have chose romantic, then it would be a lot slower and it would have given you much more in depth descriptions about who was gazing into whose eyes and what kind of emotions they wear on their faces or whatever. Let's tab.
Speaker 3:
020/030 left right slider, 67%.
Robin Christopherson:
So that's the progress through the video. Tab again.
Speaker 3:
Rewind 5S button. Forward 5S button. Alt plus tab. Vol: left right slider.
Robin Christopherson:
That's the volume, tab again.
Speaker 3:
Current audio description: read only edit.
Robin Christopherson:
So this is the pane where the audio descriptions pop up in real time and are voiced by your screen reader. Tab again.
Speaker 3:
AI token usage: read only edit.
Robin Christopherson:
So this is a panel about the AI token usage. Now I'm on the free tier and I've done lots of videos and it hasn't brought up any kind of error. I don't know how many million tokens are used each time, it's a big number, but it hasn't complained yet. If I tab again.
Speaker 3:
Explore scene... Button.
Robin Christopherson:
So this is really cool, explore the scene. So basically if I hit tab enter on this or space bar, it would bring up a picture of the current position in the film, in the video, like the frame, and it would put a sort of grid up on the screen. And you can arrow up, down, left, and right through it, and it will speak the objects that are where you are. So you can kind of get a really good idea of the layout of the frame, which objects are where. Super cool. If I tab.
Speaker 3:
Ask more... Button.
Robin Christopherson:
Ask more, that's your typical Ask AI more about this scene. Well, this snapshot of the video. Tab again.
Speaker 3:
Edit descriptions... Button.
Robin Christopherson:
So here, if I hit space or enter on this, I could edit any descriptions I wasn't happy with before they get burnt into the MP3. Again, super cool. I say MP3, but you can actually export it as a SRT file, which we'll talk about in a second as well. Tab again.
Speaker 3:
Settings... Button.
Robin Christopherson:
Settings, that's the same settings panel as we've been through before. Tab.
Speaker 3:
Export... Button.
Robin Christopherson:
Export. Let me hit space on that.
Speaker 3:
Video player with descriptions dialogue.
Robin Christopherson:
First of three choices.
Speaker 3:
Continue button.
Robin Christopherson:
Oh, we're on the continue button. Let's tab.
Speaker 3:
Cancel button.
Robin Christopherson:
And tab again.
Speaker 3:
Select export format MP3 audio left paren descriptions mixed with original video right paren radio button checked. 1 of 3.
Robin Christopherson:
Okay. So descriptions mixed in with the original video audio and that'll be an MP3 file. If I arrow down.
Speaker 3:
Select export format SRT subtitle file left paren descriptions of subtitles, right paren radio button checked. 2 of 3.
Robin Christopherson:
So descriptions are subtitles. So in YouTube, if you've got a screen reader like Voiceover on iOS running, then it will speak out the subtitles, for example, any captions. And this is exactly what will happen here. The .SRT file brings up text at the right time in a video, and in this case it would describe what the action is and it would be voiced by your screen reader. If I arrow down.
Speaker 3:
Select export format plain text file, left paren .text right paren left paren descriptions line dash by dash line. Right paren radio button checked. 3 of 3.
Robin Christopherson:
This is really cool. It's just a text file, and on each line is the audio description. There's no timestamps or anything. So if you just want an overview of a video, just the action without any dialogue or anything, then just export it as this text file and just arrow down and listen to everything that happens visually in the video. If I arrow down.
Speaker 3:
Select export format MP3 audio.
Robin Christopherson:
Back up to the top. Okay, so let's tab back to the...
Speaker 3:
Continue button.
Robin Christopherson:
And hit space or enter.
Speaker 3:
Save exported file dialogue. File name: edit combo. Repackaged underlying C5L37.
Robin Christopherson:
So it's asking what the name of the MP3 file should be. It gives it a weird name. I don't know, based on the downloaded video, I'm not sure. So let's type Oakley smart glasses and enter and it will save it.
Speaker 3:
Omni Describer-V1.
Robin Christopherson:
In a folder.
Speaker 3:
Video player with descriptions dialogue. Export... Button. Cancel button. C:\ okay button.
Robin Christopherson:
Okay, so it's done that. And hit space.
Speaker 3:
Omni Describer-
Robin Christopherson:
Okay. Let's go to the folder where that file is.
Speaker 3:
Oakleysmartglasses.mp3, 32 of 45.
Robin Christopherson:
Okay. Here we are, let's hit enter. And we'll hear those audio descriptions in between the dialogue, hopefully not interrupting each other and in the SAPI 5 voice this time, and ducking the audio behind it every time they speak.
Speaker 3:
VL's menu.
Speaker 6:
Johnson, skate JR Smith his shot wear skates. Bicho Carrera, Paraglider.
Speaker 4:
Play my hooping playlist.
Speaker 6:
J.R. Smith dunks. Shot wear Skates Bicho Carrera paraglides.
Speaker 5:
Hey Meta, take a video.
Speaker 6:
Skater jump stairs.
Speaker 5:
Hey Meta, how strong is the wind today?
Speaker 7:
The wind is blowing 12 miles per hour.
Speaker 6:
J.R. Smith golfs. He lines up put. Ball sinks. Oakley Meta logos.
Robin Christopherson:
Wow. I think one was overlapped or cut off, but it did its absolute best to give really short, succinct audio descriptions there. Wow. And see, it gave everyone a name. Where did it get those from? How does it know who those people are? I don't know.
So there you go. And obviously if this was a different kind of video, a much slower moving one, then you could choose a different profile for the model and you'd get much more in depth, longer descriptions about the action. But yeah, hopefully this is, I mean, it's quite an extreme example, but I thought it did really well. At least I know what's happening in the video now.
Huge thanks to Omar for creating this free and amazingly useful utility. Omni Describer, if you just search for that, you can get straight to the page where you can download it. And he's really responsive. He squashed a bug that I spotted and he did it within 24 hours. So huge shout out to Omar there. And yeah, tell your friends guys, Omni Describer looks pretty cool.
Jonathan Mosen:
Indeed. Thank you, Robin Christopherson, for that demo of Omni Describer. It's certainly something that a lot of people on Mastodon have been talking about for a while because of how well it works. You can follow Access On on Mastodon, along with other NFB related accounts. In fact, we're at [email protected]. [email protected] on Mastodon.
Let's go to Christopher Wright, who says I was listening to the episode about the Brava product, and I'm very impressed. As far as I know, this is the first smart home manufacturer that's taken screen reader support and accessibility seriously.
I hope they release a model that's fully accessible using the interface on the device for a couple of reasons. The first is obvious, we should have native access just like anyone else. It would allow us to independently use the product we paid for right after taking it out of the box.
The second is inevitable planned obsolescence. We saw it with Google Nest thermostats and Wemo discontinued all internet connectivity from some of their smart plugs in February 2026. This isn't a big problem for sighted people, but the Nest was turned into an expensive paperweight for blind people, and I'm afraid that's what's going to happen to my amazing oven I paid over 1,000 dollars for in a matter of five or six years.
Or because the manufacturer got tired of supporting the software and wants people to upgrade to another newer model that may or may not be better than what I already have. Why should I replace perfectly good and importantly, crucial hardware that still technically works just because of that?
I'm watching the space very closely. We definitely live in interesting times. If Brava or any other manufacturer creates an accessible interface on the device itself that will continue working, even after the plug is pulled for the smartphone app, I'll seriously consider picking it up.
Thanks, Christopher. I'm certainly enjoying my Brava smart oven at home, and I see other people on social media who've purchased them are enjoying theirs as well.
Now let's go to Rich Yamamoto who says, "I was wondering if it was possible to connect my Windows PC to an Android TV wirelessly through either casting or screen mirroring." I have an 80 plus inch Clear Touch TV in my classroom, and I use it to post lesson information, videos, and whatever else I feel is useful for posting.
I know that I can mirror using my Mac, but my job is primarily Windows based. I could do it through HDMI, but then I would have my back to my students, and I don't really like doing that if I can avoid it. Also, when I cast or mirror, how do I select which window goes on the TV/projector? I want to be able to move something over to the TV and still be able to alt tab around the system to reference other documents that I need to also have up on my computer.
Thanks very much for writing in, Rich. I think this is all possible because Windows has wireless display support built right in, and it works with Android TV through a standard that's called Miracast. And most Android-based displays, including the Clear Touch panels that are commonly used in classrooms like yours do support Miracast. And just to be clear, for those listening and not reading this, it's spelled M-I-R-A-C-A-S-T.
You don't need to install anything extra on the Windows side. If you want to connect, you press the Windows Key + K and that will open the cast panel. This is a standard feature of Windows, so it is fully accessible.
And you should be able to navigate and find the Clear Touch display as a device that you can use to connect to, assuming that both are on the same WiFi network. So that is important that you and the TV are using the same network. You then select it and Windows will establish the wireless connection.
When you connect to that wireless display, Windows will give you the option to either duplicate your screen or extend it, and you want to choose to extend your screen. And you'll want to extend it because this turns your Clear Touch into a second monitor.
Whatever you drag over to that monitor will appear on the big screen while your laptop screen stays completely independent of the Clear Touch. And you can then alt tab freely on your laptop and look at your notes or check other documents, whatever you need to do, and your students only see what you've deliberately placed on the TV.
Once you're in that extend mode, you have to drag the window to the Clear Touch, and you can do that from the keyboard, I'm pleased to say. Just press Windows Shift and the left or right arrow key to move the focused windows between monitors, and that should get the job done for you.
And there's even a plan B if you find that Miracast support just isn't reliable on your school's network, and that can happen from time to time. Another option that you have is to look at whether your Clear Touch has a built-in Chromecast or supports Google Cast.
And if it does, you could cast a Chrome browser tab from your Windows PC, which is more limited, of course, than full screen extension, but it works well for displaying web-based content and videos, and sometimes it's just a bit more robust. So I hope this gets you up and running.
And here's Hayo Simkin who says, "I'm considering signing up for Preply Tutoring," that is spelled P-R-E-P-L-Y Tutoring, given my proficiency in English, Hebrew, and Arabic. Could you or the listeners provide information on the platform's accessibility, including the app? I'm interested in the entire process from registration to booking sessions. I understand Preply requires a headshot and video submission, but I can enlist a friend's help.
Thanks for writing in. I've not encountered this platform before, but I have done a quick bit of research and found out what I could find out. Preply does have an accessibility help page, but frankly, it is quite thin.
And on it says that it is committed to creating a great experience for all students and tutors, and that it is researching how people with disabilities use screen access technology on its website and app. So it's not saying that it's accessible right now, just that it's researching it and finding out more. So if you do sign up, sounds like you could be a guinea pig for them potentially.
But really it's pretty basic. It talks about turning on voiceover at the Mac system level, which of course has nothing to do with Preply specifically and how to use Browser Zoom. That's a pretty low bar.
It looks like for registration, the process currently requires you to register as a tutor only in the desktop version of Preply. And it involves filling out an eight section application that includes a professional looking profile photo, certifications, education credentials, a profile description, and a tutoring rate. The application is then reviewed within five business days.
And you're right when you say that a headshot is required. That is something in this day and age that you could do yourself with the various ways of positioning your face in the frame and then verifying the photo with an AI service. But if you feel more comfortable, you can, of course, use a sighted person as an accommodation to assist you to get that part of the process done.
But there are other visual elements to this as well. There is a two-minute welcome video that is described as a must for any Preply profile. And a profile can be hidden from search results if the introductory video can't be viewed to the tutor's profile. So this isn't optional in any practical sense.
So there may be others listening who've tried this platform, Preply, and can give us some firsthand account of how well it works. All I could really glean from taking a look at the site was they clearly know what a screen reader is, and the fact that they're not oblivious and that they care enough to have that page is a good sign.
Just how well it works in practice, I guess, remains to be seen. And if you do choose to try it, let us know how you get on. We will have more listener contributions in a moment as Access On continues from the National Federation of the Blind.
Speaker 8:
For over 100 years, the American Action Fund for Blind Children and Adults has provided programs and services to the blind and deafblind, mostly without cost and always with an emphasis on Braille. Thousands of Braille books and Braille calendars sent and thousands more blind and deafblind people to reach. Help the American Action Fund by contributing today at actionfund.org/donate. Thank you for your support.
Speaker 9:
Advocacy doesn't end with awareness. It ends with access. Under ADA Title II, public entities must ensure their digital content is accessible by April 2026. That includes existing documents people rely on every day. Many agencies underestimate the scope, thousands of inaccessible documents, limited time, limited budgets. The good news is that large scale remediation is possible. For clear guidance on what the rule actually requires, visit www.title2.info.
This message is brought to you by Numa Solutions. We've remediated hundreds of thousands of pages in days, aligned with WCAG 2AA guidelines at enterprise scale and speed without sacrificing usability for blind users. Accessibility isn't a privilege, it's a right.
Jonathan Mosen:
This is Access On from the National Federation of the Blind. Let's continue looking through some listener emails, and Otto's writing in, he says, "Hi, Jonathan. It's nice to finally have made the jump from Living Blindfully to Access On. I missed the announcement on Living Blindfully, and now I'm catching up. The presentations of .PADx and Monarch have been interesting, but they can't do what I'm looking for, at least yet.
I found the Graphiti a couple of years ago, but because I live in Sweden, I haven't been able to get my hands-on one to try it myself. I'm a developer, and it would be extremely helpful to be able to get a feel of the visual layout of my applications and webpages. The Graphiti page states that it works as a tactile display when you connect a video output to it, and there's some videos showing any clips as it happens.
This is not the same as being able to explore the screen of a Windows computer. And I wonder if you or someone at NFB have tried the Graphiti in this use case. It costs 15,000 American dollars, so it's critical to confirm that it's possible to do what I want before spending all that money. If you can't enlighten me, perhaps you know someone who can.
Otto, I can't enlighten you because at the moment our Graphiti is in for servicing, but yes, I do know exactly the person who can. And that is non other than Venkatesh Chari, who is the chief executive officer of Orbit Research. I forwarded your email to him for his response, and this is what he has come back with.
He says, "Yes, absolutely. The Graphiti and Graphiti Plus can function as a tactile monitor and display, whatever source you connect to the HDMI input port on the devices. So you can simply connect an HDMI cable from the computer to the Graphiti, and it'll display your Windows or Mac or Linux screen in real time. It is plug and play with no software or drivers required.
At conventions, we often demo this with an iPhone connected to the Graphiti and browsing the booth visitor's favorite webpage. We also show the camera app on the phone, and this provides a live tactile view of the camera on the Graphiti. We are also working on software that allows you to control the Windows pointer with your finger just as though you are using a touchscreen. We have a prototype of this working on the Graphiti."
So that's all very cool, isn't it? So thanks for raising that question, Otto. We're looking forward to getting our Graphiti back at the International Braille and Technology Center. There are certainly other devices, some of which we talked about last week in our CSUN recap. So we will have plenty more to say in the coming years and months about multi-line Braille and graphical displays.
Here's Mason who says, "I'm emailing in from the newly launched Orange County chapter here in Orange County, California, and thought of emailing regarding my Waymo ride experience." Well, I was in Orange County not so long ago, so marvelous to hear from you, Mason, marvelous.
He says, "I have ridden the service nine times, two times independently and the rest with other blind and low vision individuals, and the experiences of each have been very positive. As a blind and low vision individual, having equal access to autonomous vehicles, as well as getting into a car independently without having the need of associating with any other individual regarding that matter is very liberating and gives quite a bit of freedom and autonomy for me in that regard.
I have been listening to your podcasts on the NFBRN and have found everything very helpful as I'm navigating a new space as a new member entirely."
Well, thank you, Mason, for writing and I really appreciate that. And you can write into, if you would like, [email protected]. If you are a member of a chapter of the National Federation of the Blind, do what Mason did and tell us where you're emailing in from and what chapter you are a part of.
I'm glad that Waymo's working out so well. We certainly have a constructive relationship with Waymo. They work with us to continually evaluate and refine the experience that we as blind people have of these Waymo vehicles.
I was in Austin recently and I really wanted to go for a ride in one of those and wasn't able to, because there's a thing that you have to choose in the Uber app that says, "I'd prefer to be matched with Waymo," but that doesn't guarantee that you're going to get one every time you request a vehicle. And my number didn't come up, but like many of us, I will be back in Austin in July, so I'm hoping my number comes up then.
All right, it's not the first time we've talked about this subject and I feel sure it won't be the last. I'll read a couple of representative messages on the subject of Facebook. Let's go to Louisville. Rick Roderick is writing in and he says, "I use Facebook with the PC and the Firefox browser. I am finding the interface increasingly difficult to use. My cursor is moving all over the place. When I hit see more, I am often taken to the top of the screen. The interface is very screen oriented. If a post is long, I hit H to go between. After two or three posts, I hear wrapping to top.
I think it is time for a new blind-friendly interface. Here are my suggestions. It should be linear and not screen oriented. Perhaps the user would be able to specify a number of posts to be shown at a time. For example, after 20 or 30 posts, see more posts would show. One should have the ability to see all posts or choose to get to see more message. The refresh rate could be slowed or stopped. For example, I might specify refresh every five minutes. I might also specify no refresh unless I hit See More Posts.
Another possibility would be being able to specify parts of the screen I don't want to see. Right now, I may be reading posts when I get a whole segment of groups I might want to join or go through people I might know. This should be expressed as links rather than being splashed on the screen. I only want to check these every few days.
I think all of these changes would make the experience more interesting for those who use Braille displays and speech on the platform.
Thank you, Rick. I will read another message on this in a moment. I will say that based on some of the prompts you're getting, it sounds like you might be using JAWS. And for whatever reason, our findings have been that while Facebook is far from perfect on the web on any platform for a screen reader user, NVDA does seem to do a lot better at the moment on the Facebook website.
If you use NVDA, which of course is free, so even if JAWS is your primary screen reader, you may want to try this and see if it performs better for you on Facebook. You can turn browse mode off and use the J and K keys to navigate from post to post, and it does seem to behave itself fairly well. There is a set of hot keys that you can use on the Facebook website that sometimes will clash with browse mode or virtual cursor commands.
Some years ago, Matt King from Facebook and I think Glen Gordon from Freedom Scientific got together and came up with the standard, which would allow you to put a bit of code on a webpage. And this would tell JAWS to exempt certain navigation quick keys from that particular page, and that allowed you to zip around the Facebook page with their built-in navigation commands.
Somewhere along the line, maybe with changes to the Facebook page, that bit of code got lost and it doesn't appear to work anymore. So certainly there is a lot that could be done to the Facebook page to improve it.
And Dennis Long says something similar. He says, "When will Facebook.com improve, I've noticed no improvement while using JAWS. I tried the site that was recommended, but it always redirects to Facebook.com."
And that certainly is a very fair and important question, and we would like to see Meta moving on this. We are concerned to note that there have been, as there have been in many other tech companies, some layoffs of late, which concern us in the sense that there may have been some institutional knowledge lost that may have been very helpful for an accessibility focused revamp of the Facebook website. So it is very concerning and clearly people's patience isn't infinite, nor should it be.
There is now a Facebook accessibility feedback form. For a while there, things kind of really got lost in the bit bucket and it was quite difficult to get a URL that we could give people that you could send feedback to Meta. Now we do have that. Unfortunately, it is so long and convoluted that I cannot include it here on the podcast.
I can't possibly verbalize it. But I will put it in the show notes as a link. So if you check the show notes for this podcast from wherever you got it, you will find a link in the show notes which includes the feedback form, and you can go to that URL and provide feedback about what you're experiencing.
Obviously, this is all in Meta's hands. We will keep advocating for as long as it takes. And I can't say that completing that feedback form is some sort of magic bullet, but I do know that it gets looked at. I do know that that form is live and it's genuinely going somewhere where it can make a difference. So if you are having experiences with Facebook, then please pass it on.
We have also received some feedback through our channels from members about recent regressions in Facebook for iOS. When we receive that feedback, we do pass it on directly through channels that we have, and we hope that it will be addressed soon.
But now that this feedback form exists and that's progress of a sort, I would certainly encourage everybody to use that feedback form about accessibility problems you are having so that there can be no doubt that Meta is aware of them.
Next email on a different subject comes from Eric Duffy. I know this because it says, "This is Eric Duffy, Treasurer of the capital chapter of the National Federation of the Blind of Ohio. I recently spoke with a student from Miami of Ohio University. She mentioned the inaccessibility of delivery robots on campus. She specifically mentioned that Grubhub uses these robots. She cannot find these delivery robots to retrieve her delivery order without sighted assistance.
It seems to me that she can't be the only blind student experiencing this problem. I feel certain that these robots are used on other campuses. I would like to know how others are dealing with them and to ask that the NFB consider educating the manufacturers of these robots about the need for non-visual accessibility if we are not already doing so."
Thanks, Eric. Let's get some feedback on this. Have you encountered these robots delivering you your food? There's got to be some way of pinging them, right? There should be a way of pinging them with the app so that they make a sound, much like the work that we've done with Waymo, one would think.
That will be a good strategy. So let us know how much of a widespread issue this is. If you're experiencing this, drop us an email, [email protected]. Regardless though, Eric, we'll certainly take a look at this, and I thank you for bringing it to our attention.
The prolific Curtis Chong is back and he says, "Hello, Jonathan. I serve as the Treasurer of the National Federation of the Blind in Computer Science. In my various roles as a member of the Federation, I routinely host quite a few Zoom meetings. A recent update to Zoom to version 6.7.5 has caused JAWS 2026 screen reader to generate excessive speech." Yes, indeed. And if you've seen this, it's not nice. It's not pretty at all.
Curtis continues, "If you have purchased the Zoom Pro scripts from Hartgen Consultancy, the latest update to these scripts turns off the successive chatter. However, for those who have not acquired these scripts, there is a relatively painless way to fix this problem, at least until the folks at Zoom implement effects in the March update to Zoom.
Here is a procedure which assumes you are using JAWS without the Zoom Pro scripts from Hartgen Consultancy.
Get into the Zoom workspace program, but do not start a meeting or anything like that. Press the JAWS key along with the number six to bring up the JAWS settings center. Tab once to focus on the tree view. Arrow down in the tree view until you reach JAWS Verbosity, which should be reported by JAWS as being closed. Press right arrow to open this. Arrow down once to hear what level is set for speech verbosity.
For most folks, this is stated as beginner. My speech verbosity is set to intermediate, but that doesn't really matter. Arrow down once to configure verbosity levels, which JAWS will report as being closed. Press right arrow to open the branch of the tree. Assuming you're at the beginning verbosity level, arrow down once to hear beginner. Do not press enter on this. Press the space bar instead. You should now be in a list of 24 items.
Press the letter H to focus on help balloon, which JAWS will report as being checked. Press space to uncheck this. Tab two and activate okay for as many times as needed in order to exit JAWS settings. This should do the trick and save you lots of aggravation during your Zoom meetings."
Thank you so much for sharing that, Curtis. And of course, once you're a quitting setting center, be sure to answer yes when you're asked if you want to save those changes that you've made.
Chris Hofstadter says, "Hi, Jonathan. As per your recommendation on Living Blindfully, I have my battery setting to not charge beyond 80%. It is an iPhone 16 with the latest software from Apple on it. Since I have bought the phone, it seems to have been working properly, but once in a while, it will charge all of the way to 100%.
For the past couple of days, since my last software update, it's been charging to 100% and I checked the setting and it's still set to 80%. This is obviously not a crisis, but I wonder if Apple is forcing some sort of AI thing onto my charging settings."
Thanks for writing in, Chris. At least some of what you're experiencing is by design and doesn't have anything to do with Apple sneaking any AI behavior into your charging settings. When you enable the 80% charging limit, Apple will occasionally allow the phone to charge all the way to 100%.
That is intentional calibration behavior and Apple has documented that. It is built in to help the iPhone accurately track your battery's health and capacity over time. Without those periodic full charges, the battery management system can lose its frame of reference, so Apple has built in those exceptions deliberately.
So that's the normal bit, it's what you've been experiencing. But I do want to acknowledge that what you're describing does sound a little more frequent now than the occasional top-up that you would be expecting. If it's been happening consistently for a couple of days or longer by the sounds of it, it is worth keeping an eye on whether it settles back down to the usual pattern. In most cases, it does, but if it continues charging to 100% every single night going forward, it would be worth checking a couple of things.
First, you want to make sure that optimized battery charging is still enabled alongside your 80% limit, because the two features work together. You can find both of those in the settings and then battery and then battery health and charging. Second, if the behavior really does seem persistent and unusual after a week or so, then yeah, I would start to get a little concerned about that and I guess you would have two options in that event.
One is to just wait until the next software update drops in case it's some sort of bug that has crept in that would get resolved. Because you'd think with something as significant as the battery, if some bug has crept in, it's likely to be one of those things that's going to get squashed quite quickly.
If that doesn't work, then maybe there's some sort of fault developing with that battery and it will be worth contacting Apple Tech support to see what they recommend. I suspect that the first thing they'd recommend is a full reset of the device. And the cure may be worse than the disease, but that's a call that you would have to make.
Francisco is writing in on his favorite subject, WhatsApp, and he says, "First, I wanted to thank you for reporting to Meta the issues I've been sharing with you and the rest of the audience. You were absolutely effective because all of them got fixed." Well, that's good. I mean, I know that people are pretty frustrated right now. We would find rather issues did not develop in the first place, of course, but the plan B is at least we can normally try and get a speedy resolution. Some quality control would be good though.
Francisco continues, "And because it works so well, here is another issue that you might want to bring to their attention. On the new WhatsApp Windows app, there were two call buttons inside a chat window, one for voice and the other for video call, but that is no longer the case. Now there is only a call button that opens a menu with audio, video, and an option to schedule a call. Well, JAWS reads the options on that menu as a single sentence, but the menu is not accessible.
With tab, arrows, and even enabling the virtual cursor, I was unable to interact and activate these menu elements in an accessible way. Would you please report this so that it is fixed ASAP? There is a workaround for this," he continues, "which is to make the call from the info button within the chat, but this is just one extra step. That call menu must be accessible and it is so easy to fix it."
Thanks, Francisco. We will report it, of course, and I would also encourage you and others affected by this to check out the link in the show notes to the Meta Accessibility Report Form, and you can also report it to them directly. I suspect, the more, the merrier.
The next email says, "This is Brian Mackey, president of the NFB of Pennsylvania Greater Burke's chapter. I also serve as the bookkeeper and web coordinator for the NFB of New Jersey and web coordinator of the Arkansas, Illinois, Iowa, Massachusetts and Nevada affiliates. NOPBC Massachusetts, Greater Springfield Chapter, and the Pennsylvania Association of Blind Merchants." My word. Thank you for all of your work for the federation, Brian.
"Like you, I started out with WordPress and prefer the classic editor for editing content. It allows me to make sure whatever I am adding to the page is formatted the way I want and guarantee that content is set with the appropriate header markup and images set with alt text descriptions. I have been using WordPress since 2015. I only use WordPress these days on a couple of sites since most of the sites I currently manage have been migrated to Drupal.
I first learned how to maintain a site in Drupal back in 2017 when I attended a web training seminar held at the national office led by the late Rachel Olivero. Last year, I was appointed to the committee on Federation website coordination and waiting to hear if I have been reappointed to the committee for 2026.
If anyone has WordPress or Drupal related questions, please contact me and I will do my best to answer them. I can be contacted at (609) 680-8488 or [email protected]."
Yasmin is writing in and says, "The city of Rockville is using the WebEx platform for their virtual meetings. In the past, I had some trouble accessing this platform. What tips and tracks can you share? I want to participate in these meetings."
Well, I think we can help there, Yasmin. If we go back to episode 26 of Access On, which was released back in May of 2025. You will hear a session on WebEx in which we demonstrate how to use it. It used to be quite problematic. It has come quite a long way in terms of accessibility in recent times. So if you had some problems maybe a few years ago, you might find it much better than you were expecting, but you can hear a demo of all of that in that episode of Access On from May in 2025.
Darrell Hilicker is back. He says, "First of all, House Hilicker from the Phoenix chapter says like Eliza Doolittle from My Fair Lady, what did she say? In Hampshire, Hereford and Hampshire, hurricanes hardly happen." I think that's what it was. Anyway, Daryl says, "I was thrilled to learn that Vispero is working on a screen reading AI agent.
I think there should ultimately be limited guardrails on this feature, as it needs to be able to accessify the world for us and we shouldn't do anything to hamper its ability to do so. And yes, that includes the ability to press that ultimate checkout, place order or submit button.
Instead of just stopping at submit, the agent should simply remind the user that an action will be taken, ask for permission to take that action, and if the user says yes, then simply get the job done without further fanfare. As a user, I know what I want and I don't need or want my screen reader to coddle or protect me from myself.
This kind of accessification is already starting, and I played around with and demonstrated a couple of early solutions. The first is a mainstream feature Google has started to roll out called AutoBrowse, AKA Chrome Gemini. The premise is amazing and straightforward. Instead of clicking and keyboarding, the user simply asks for what they want and the agent clicks the button, enters the text and selects from the inaccessible combo boxes. Is it perfect? Of course not. But is any technology?
The second approach is a blindness specific solution called Guide. You can visit guideinteraction.com, that is all one word, guideinteraction.com, to learn more about this one. It does something similar to AutoBrowse except that it sends screenshots back to the AI, which are analyzed and the best action is determined and done right on your computer.
This one has a small subscription fee, which I believe has gone up to 15 dollars a month. Guide's use of screenshots means it's much slower than Autobrowse, which of course has direct access to the DOM document object model, but it can be used in many other screen reading scenarios far beyond just web.
I demonstrated using Guide to help me drag a headline component from the component toolbox to the Canvas using Salesforce's inaccessible experience builder. Please see that article, and you can find these both on blindaccessjournal.com.
And third and finally", says Darrell, "We have Viewpoint, which is more of what many might be thinking of as a more traditional screen reader assistant. Viewpoint can, for instance, expose all the clickable elements of any otherwise inaccessible user interface so that the user can find and choose or select them by pressing tab, shift tab and arrows and enter all the space bar. It's available at no cost except the usage of your own Gemini API key.
Let's face it, there are lots of headwinds against improving accessibility, especially here in the United States. If we can use AI tools to help us accessify the world, then at this point, let's get beyond the dogma and the drama and just get that job done already."
Well, thank you very much, Darrell. And if you have any comments on that email, [email protected] if you want to be in touch. It's a really interesting time to be thinking about these things because we are certainly confronted not just with interesting technological opportunities and potential, but also philosophical considerations as well. And I, for one, find this such an interesting time because of that.
Well, it's been a while since we've done a tech tip. Let's make sure we get one in. Carl Smith says, "This tip may be obvious to some, but not for others. When you find yourself on a webpage or often a PDF document with information you want to copy, but find that when you try copying with the virtual cursor, your results are not what you expected to copy, here's a solution for you. One, press the JAWS key plus space followed by Shift H. This will clear the JAWS speech history.
Two, next have JAWS read the information you want to copy. Three, press JAWS key plus space followed by H. This will bring up the JAWS speech history in the virtual viewer, which contains the most recent information spoken by JAWS. Four, navigate the information with your virtual cursor and copy any or all of it with Control+C. Five, go back to the place where you want to paste the information and do so by pressing Control V."
Thank you, Carl. And there are speech history features and other screen readers as well, so adapting the keystrokes should enable you to try that trick with whatever screen reader you happen to be using.
That concludes this episode of Access On, the technology podcast of the National Federation of the Blind. To send in a contribution for a future episode, email us, attach an audio clip or just write it down and send it to [email protected]. That's [email protected]. To keep up to date with Access On, follow us on Mastodon. [email protected]. That's [email protected] on Mastodon.
To subscribe to an announcement only email list about upcoming episodes, send a blank message to [email protected]. That's accesson-announce-subscribe at nfbnet.org. To learn more about the National Federation of the Blind, visit our website, NFB.org or phone us, (410) 659-9314. That's (410) 659-9314, and be sure to check out the Nation's blind podcast right from where you heard this podcast.