This is really interesting. I wonder–would it be possible to listen to an audiobook or PDF at 800 wpm once one learns how to understand the screenreader "language"? Presumably the cognitive load would get heavy if the content were a stream of unstructured prose as opposed to code.
I worked with a developer briefly who would produce code at an extremely high speed (this was before LLMs) and I've observed them write 50% of the code for two projects in the matter of a few days.
While I never got around to asking them how they coded so fast, this was probably one of the tools in their arsenal.
If you compare different languages, the speed people tend to speak (measured in syllables per second) varies significantly. However, the number of possible syllables also varies significantly. Once you account for that, the speed of speaking in terms of information is fairly consistent across languages.
I'm not aware of any specific research directly on point to what the author of the posted blog describes. But his hypothesis that having a consistent speaker reduces the cognitive overhead of decoding seems to be part of the story.
However, we would expect a similar effect in people who read, as the writing is also highly standardized. However, I've generally seen silent reading speads for English estimated at around 250. Getting up to 800 WPM puts you well within the realm of speed reading territory.
The relatively high structure of code and rote emails probably helps too.
ok, so you can understand the words at 800wpm...can you really comprehend what's being said? When I listen to youtube videos at 2x speed I can usually pick apart all the words just fine but I often have to slow it down to properly process the meaning behind those words.
Just FYI OP (assuming OP is the author of the post) there's no margin on your blog. Text goes all the way to the edges.
As a related sidenote, I wonder how quickly ChatGPT replaces much of the customized tools here? ChatGPT is probably pretty proficient at being able to describe the contents of eg a screenshare, or a screenshot of a website.
i've always wondered about the possibility of hybrid visual-audio interfaces for sighted users. a screen reader producing lots of words at the same time that you're looking around the screen—perhaps the visual ui would have to privilege symbols over words, to not force you to perform two language processing tasks at once
After read about the poor scenario where the Linux accessibility tools is today (https://fireborn.mataroa.blog/blog/i-want-to-love-linux-it-d...), I was wondering: if maybe the developers start to use these accessibility tools to improve their speed reading (and productivity as well), this could also helps to prioritize the accessibility features and bug fixes in Gnome, KDE, Qt, etc.
I'm getting intense deja vu reading this. I remember a very similar article from 3-5 years ago. Even the title feels like I've seen it before. I thought this was a repost from that one but I remember the previous post had a photo of the author's desk with a keyboard, desktop, and no monitor.
> Windows might not be trendy among developers, but it’s where accessibility works best. I don’t have to worry about whether I can get audio working reliably. The NVDA screen reader works on Windows and is free and open source, actively maintained, and designed by people who are screen reader users themselves.
> That said, I’m not actually developing on Windows in the traditional sense. WSL2 gives me a full Linux environment where I can run Docker containers, use familiar command-line tools, and run the same scripts and tools that my colleagues use. Windows is just the accessibility layer on top of my real development environment.
> I use VS Code. Microsoft has made accessibility a core engineering priority, treating accessibility bugs with the same urgency as bugs affecting visual rendering. The VS Code team regularly engages with screen reader users, and it shows in the experience.
MS has to make accessibility a priority because it's mandated by government (its customer).
Smaller companies would benefit from better libraries and design systems that make it easier to incorporate accessibility. Make accessible the default.
I was wondering what TTS voices you use? I've heard from other blind people that they tend to prefer the classic, robotic voices rather than modern ML-enhanced voices. Is that true in your experience, too?
The interesting part for me was that you can recognize synthetic voice much faster than human speech. Is there a specific voice you are using for 800wpm or it can be any TTS? Also, I think older voices sound more robotic that the newer ones (I mean pre AI, like the default on android is newer for me). Is there a difference for how fast you can listen to the newer more nicely sounding ones or the older more robotic ones?
This gave me new perspective. Both on what's possible with our ears/cognition and what our eyes do for us (for example seeing the extra quote so easily). Appreciate this very much and Thank you for what you do Neurrone!
For anyone wanting to trying this out on YouTube, select your video, open the Web Developer Tools with Ctrl+Shift+I, then type this (followed by a carriage-return):
document.getElementsByClassName("video-stream html5-main-video")[0].playbackRate = 5; // Or whatever speed you choose
22 comments
[ 2.7 ms ] story [ 59.7 ms ] threadWhile I never got around to asking them how they coded so fast, this was probably one of the tools in their arsenal.
If you compare different languages, the speed people tend to speak (measured in syllables per second) varies significantly. However, the number of possible syllables also varies significantly. Once you account for that, the speed of speaking in terms of information is fairly consistent across languages.
I'm not aware of any specific research directly on point to what the author of the posted blog describes. But his hypothesis that having a consistent speaker reduces the cognitive overhead of decoding seems to be part of the story.
However, we would expect a similar effect in people who read, as the writing is also highly standardized. However, I've generally seen silent reading speads for English estimated at around 250. Getting up to 800 WPM puts you well within the realm of speed reading territory.
The relatively high structure of code and rote emails probably helps too.
As a related sidenote, I wonder how quickly ChatGPT replaces much of the customized tools here? ChatGPT is probably pretty proficient at being able to describe the contents of eg a screenshare, or a screenshot of a website.
"I use my computer during natural pauses in the conversation or presentation."
Casually dropping that everyone is speaking so slow. That you must use the time between sentences for something meaningful, pretty funny. :-)
imagine the bandwidth
After read about the poor scenario where the Linux accessibility tools is today (https://fireborn.mataroa.blog/blog/i-want-to-love-linux-it-d...), I was wondering: if maybe the developers start to use these accessibility tools to improve their speed reading (and productivity as well), this could also helps to prioritize the accessibility features and bug fixes in Gnome, KDE, Qt, etc.
I tried searching algolia, but I can't find it.
> Windows might not be trendy among developers, but it’s where accessibility works best. I don’t have to worry about whether I can get audio working reliably. The NVDA screen reader works on Windows and is free and open source, actively maintained, and designed by people who are screen reader users themselves.
> That said, I’m not actually developing on Windows in the traditional sense. WSL2 gives me a full Linux environment where I can run Docker containers, use familiar command-line tools, and run the same scripts and tools that my colleagues use. Windows is just the accessibility layer on top of my real development environment.
> I use VS Code. Microsoft has made accessibility a core engineering priority, treating accessibility bugs with the same urgency as bugs affecting visual rendering. The VS Code team regularly engages with screen reader users, and it shows in the experience.
Smaller companies would benefit from better libraries and design systems that make it easier to incorporate accessibility. Make accessible the default.
Happy to answer any questions.
I was wondering what TTS voices you use? I've heard from other blind people that they tend to prefer the classic, robotic voices rather than modern ML-enhanced voices. Is that true in your experience, too?
The interesting part for me was that you can recognize synthetic voice much faster than human speech. Is there a specific voice you are using for 800wpm or it can be any TTS? Also, I think older voices sound more robotic that the newer ones (I mean pre AI, like the default on android is newer for me). Is there a difference for how fast you can listen to the newer more nicely sounding ones or the older more robotic ones?
There were browser extensions that would present web content at 800 WPM, and they were fantastic. I wish they hadn't disappeared.