RobMurray

It's still AT-SPI for wayland, the main difference is how screen readers grab keyboard input events.[0] I don't think there is a big difference from a toolkit point of view. I don't personally have experience with Wayland because most blind people recommend Mate as being the most accessible desktop still.

Thanks for considering a11y for your toolkit - it really makes a difference to those of us who are disabled. Are you implementing a11y separately for each platform? If you use accesskit[1] you only have to implement it once for all platforms. I recently vibe coded accessibility for the Swell toolkit[2] used by Reaper. I have a branch using accesskit and a branch implementing at-spi. Accesskit made things a lot easier and more performant.

Let me know if you would like a screen reader user to help with testing your toolkit.

[0] https://lwn.net/Articles/1025127/

[1] https://github.com/AccessKit/accesskit

[2] https://github.com/RDMurray/WDL/tree/accesskit

and my fork of accesskit with some features and fixes for unix: https://github.com/RDMurray/accesskit/tree/swell-fixes

There is a factory reset procedure, I think you hold the shutter button while switching them off and on again.

Most of those videos are chatGPT voice mode, which still used gpt 4o last time I checked. it is far from SOTA.

Accessible PDFs are quite rare in reality. Especially if there are tables, graphics, maths, forms or anything more than plane text.

Apple's stt has been on-device for a long time now, long before iPhone 16. I haven't noticed any improvements since my first ever iphone 5S. I'm pretty sure wispr flow can use on-device models. I use Voiceink[0] which can use parakeet models on-device and can optionally use cloud models.It's like night and day comparing Apple's to Voiceink. The only advantage I find to Apple's stt is less friction. 3rd party apps just can't integrate as smoothly with the system. There's a gesture to activate Appledictation when Voiceover is on.

for every person like your coworker, there are probably several who have a much harder time with technology and who would benefit from a simpler interface.

If this includes improvements to the screen recognition feature in Voice Over, it could provide accessibility for apps where the developer doesn't care about accessibility, which is extremely common.

The vision capabilities could be useful if they are done well, but I suspect that will always be covered better by 3rd party apps.

I know plenty of blind people who have their voice speed unbearably slow and barely scratch the surface of what technology can do for them. I think an interface where you can tell your phone what to do in natural language will really help a lot of less technical people.

I'm not getting my hopes up though given apple's history with Siri, which is truly awful.

not just lan mode, you have to enable developer mode, which blocks cloud access entirely so you lose advertised functionality.

I'm working on a visual description app for the blind. even Gemma 4 E2B can give very useful image descriptions while at the same time taking questions as audio. It's also much faster than most of the current popular cloud based apps like Be My Eyes.

why? it's mostly reads. the weights are static.

First time I have seen AI psychosis being used in this way to apparently describe something positive. I think AI psychosis is a real problem that we have only just started to scratch the surface of.

Why is floating ground still a thing? I've found it can actually sting quite a bit if you are grounded or touch something that is.

it's unmistakably AI as soon as the vocals come in. maybe your ears are full of shit?

I got codex to vibe reverse engineer two devices from rom dumps recently - a talking timer that uses an 8051 cpu and a custom 5 bit audio format, and an ice cream van chime box that used a z80 and a ym2149 sound chip. Quite simple devices, but it did a great job. also made a web-based emulator for both. apparently WASM is hard, but I didn't notice.

It's still Siri. Apple intelligence plus ChatGPT really didn't improve it. The speech to text is still as bad as ever. If you want to use it with ChatGPT the best way is "Siri open ChatGPT voice" which worked fine on my iPhone 12.

I will be (pleasantly) shocked if this is any improvement. They've been talking about Apple Intelligence, App intents etc for years with very little actual progress.

I don't like the sound of that. Why do humans always need to spoil new advancements by finding the worst use cases?

any headphones or speaker with skip buttons. in podcasts it skips forward 30 or back 15 seconds. perfect for skipping ads.

strawberry isn't a trick question. llms jus don't sea letters like that. I just asked chatgpt how many Rs are in "Air Fryer" and it said two, one in air and one in fryer.

I do think it can be useful though that these errors still exist. They can break the spell for some who believe models are conscious or actually possess human intelligence.

Of course there will always be people who become defensive on behalf of the models as if they are intelligent but on the spectrum and that we are just asking the wrong questions.

for a laugh enter nonsense at https://gradium.ai/

You get all kinds of weird noises and random words. Jack is often apologetic about the problem you are having with the Hyperion xt5000 smart hub.

And in copilot.

How does this compare to Happy Coder? https://github.com/slopus/happy

How have you found performance with chrome in 8gB? Is the ram shared with the GPU?

I had a look and it does look like a great deal for the price. It would probably be good for running home assistant. If you marketed a version with HA preloaded you could beat the Nexus1 AI base station [0] to market.

The thing Home Assistant needs, or any AI assistant like Openclaw, is an affordable smart speaker with a good microphone array and open firmware. I think there would be a good market for that combination, even with people who would be put off by the security and privacy concerns of Openclaw.

Yes, that makes perfect sense. Apologies for the overly negative tone of my comment.

Do you have a description of the device with specs, or is it in earlier development than that?

Are you using the technology from DotPad, or do you have your own?

Does it have a Braille keyboard for input? I would consider that essential for education.

I fully agree that education is the most important thing. Without it Braille will die. In the UK blind schools are dying out, replaced by specialist teachers who visit mainstream schools. It's just not the same and children don't have much incentive to learn Braille.

It would be good if it used interval names rather than relying on absolute notes. Eg minor third, fifth etc. Also notes played together and more and more complex chords as level gets higher. keyboard shortcuts would be great too. And a fully hands-free mode with voice input for practise while doing other things.

about power usage: sure there exists a light bulb such that this draws less power, but the vast majority of light bulbs are far less than 15W. Also the mac mini sips power when idle at only 2-4W. This isn't a criticism of the product, only the marketing material.

The other thing is that moltbot/openclaw or what ever it's name is today is massively hyped AI slop that will be replaced by the next over hyped AI thing in a matter of weeks.

Also a lot of what you learn is how to work around limitations of today's models and agent frameworks. That will all change, and I imagine things like skills and subagents will just be an internal detail that you don't need to know about.

My feeling is that focusing on the display hardware is the way to go. I have seen so many great Braille devices let down by poor, slow, out-of-date software, especially since they started using Android. The computing device part will always age worse than the display. It is just sad to see a $5800 Android tablet stuck on Android 8.1 and having worse speech responsiveness than devices from the 80s, and that is not an exaggeration.

We all have a supercomputer in our pockets that can quickly connect to a Braille display via Bluetooth.

Minimal software for note taking etc is great though, but a bloated OS like Android or Windows just gets in the way and is quickly outdated.

I'm curious why you think voice input is an important feature. Does that come from the needs of blind people you have spoken to? It seems like another thing to distract from producing an affordable Braille device.

I'm really not trying to be unduly critical, but this sentence stands out as utterly baffling: "Real-time voice-to-braille pipeline converts any digital text instantly."

There are some, in particular the orbit reader[0] is much cheaper than a piezoelectric display. The trade off is that is is relatively slow to refresh and quite noisy.

There is also the dot pad[1] which is much more like a screen with a rectangle of cells that can show Braille and graphics! It is a different technology using electromagnetic actuators with latching. It can only refresh when not being touched. It's also out of the price range of most consumers, but apparently the technology scales very well so they expect the price to fall. It is also modular so users can easily replace broken cells.

The Monarch[2] is based on Dot Pad technology and also runs Android and Humanware's Keysoft software like the BrailleNotes.

[0] https://www.dotincorp.com/en/product/dotpadx

[1] https://www.dotincorp.com/en/product/dotpadx

[2] https://www.aph.org/product/monarch/

Wow your ability to research products is truly inspiring! We should campaign to get this site renamed to buyer news to reflect what really matters in life.