Hehe don't worry, I didn't forget!
Quoting the original post below, then I'll answer.
flatrute wrote: ↑Wed Dec 28, 2022 12:18 pm
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
Every time someone asks me this question, they always apologize as if it's actually rude to ask. But it's not! I wish more people took the time to ask, because hopefully explaining it helps
improve assistive technology and give better insight.
I was just mimicking what white English speakers would say
If I were able to use my mother tongue I would just be normall formal. Anyway thanks for letting me know you are fine with such questions.
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
To answer though, I should clarify I'm not fully blind. I still have a decent amount of vision, enough to navigate. I just have Retinitis-Pigmentosa so I lose vision over time as I age, and I also have a few other conditions secondary to it. The worst being a severe astigmatism that makes it borderline impossible to read anything more than a few words a second.
I forgot blindness is actually a spectrum. I myself have a quite bad nearsightedness even before I attended to primary school and my mom just thinks I am "practically blind" when I cannot tell details of her face without my glasses
Also I searched the name and damn it is sad just to think about being born with sight losses over time
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
As far as assistive tech, I don't like bloatware on my system. If the accessibility tools built into my OS aren't usable then I just won't use that OS. If they're usable but missing creature-comforts then I'll program my own tools.
Making a program to assist oneself is just so cool. I would like to do something similar by making a true
IME for Vietnamese languages since existing softwares here feel too hacky to me.
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
Usually I'll use the system's built-in zoom feature - Windows has Magnifier, Macs have zoom, and both Kwin and Compiz have usable zoom plugins on Linux. I try to keep as many programs and websites in dark mode as possible, but will use color inversion where that's not possible. Dark Reader is extremely useful for this too.
Are there any good benefit for having everything in dark mode for you? I only know that most normalsighted people have dark mode for things like saving battery life on devices with OLED screens and because they usually live in a (badly lit) bedroom
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
When it comes to reading, it depends on what I'm doing.
- Browsing the Internet: I use a Chrome extension that lets me select text, right-click it, and read it aloud. It uses Google's cloud TTS voice and tries to detect the language of the text, and I hate this. My assistive tech should NEVER stop working because of slow/non-existent internet, and it should only ever read in English...my native language. Every now and then it likes to read source code in Russian, Japanese, Chinese, or something else unintelligible to me. Chrome extension devs, will ya stop?
- Discord: Right-click a message, hit "Speak Message." Only thing that can't handle is spoilers, embeds, and images.
- Programming: I use the JetBrains IDEs and I wrote my own plugin for doing exactly the same thing my Chrome extension does. Highlight code, right-click it, read it aloud using the system's TTS voice (Narrator on Windows, VoiceOver on Mac, and espeak on Linux). This plugin isn't open-source since it's tied specifically to my computer and is annoying as hell to initially set up, but it works.
It sounds like a wild west to me. Sure, they are fundamentally different programs doing different tasks but I expect that one could just use bundled screen reader to read out loud most of the text. Also Internet dependency is bad but the lack of metadata for easier processing in most contents is worse in my opnion.
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
I don't use active screen readers (that is, screen readers that read everything on-screen) because I still have a decent amount of eyesight and can generally find my way around UIs using context clues (colors, standard layouts, etc.) I rather only have text read to me when I need it.
What voices do you use when you need to have texts read?
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
At school I'd always scan assignments in as PDFs and write over them using OneNote. It was not at all ideal, because the scanner was really low-quality and half the time I was being given faded photocopies to begin with. Life's a lot easier now that I'm out of school.
Before attending higher education teachers in my schools just mass photocopied their courses and assignments and gave them to each students in their classes since textbooks are kinda sucky for them for some reason
Even now I just had to camerashoot my answers when I have to write them on papers and send them on Moodle of my college without having to scan anything.
acidiclight wrote: ↑Wed Dec 28, 2022 4:16 am
When playing games, I try to stick to console since controls are more standardized. Modern consoles have accessibility tools which is cool, but I still use a PS3 which doesn't. So I play all of my consoles through my capture card using an OBS preview window as the screen and Windows Listen mode to have the audio play through my headset. This lets me use the Windows magnifier and color inversion. If I really need to, I can also easily screenshare the game to a friend on Discord and have them help me out.
What other consoles do you have?
Edit: I actually have troubles using the right words, even in my L1.
Are there any good benefit for having everything in dark mode for you? I only know that most normalsighted people have dark mode for things like saving battery life on devices with OLED screens and because they usually live in a (badly lit) bedroom
Absolutely. In fact it's mandatory. Back in the day, before dark themes were common, I had to invert colors so text would be bright and the background would be black. While this definitely worked, it has its issues. Pictures/photos would get inverted, leading to...well...uhhh... gonna pick on Matt since it's his forum, but...this absolute nightmare of a hellspawn.
As you can no-doubt imagine, that isn't exactly a pleasant experience no matter how much ya enjoy MattKC's content. So running everything in dark mode makes sure that pictures appear normally.
The reason I can't deal with light mode isn't because it particularly hurts my eyes, it's just that I'm so sensitive to bright light that...well...it both hurts my eyes and glows over the text. So I may as well be looking at a glowing white rectangle, as if you took the polarizing filter off of your screen and there truly was nothing to see on it. The font would otherwise need to be so bold and huge that a single word would take up at least half of the screen.
It sounds like a wild west to me. Sure, they are fundamentally different programs doing different tasks but I expect that one could just use bundled screen reader to read out loud most of the text. Also Internet dependency is bad but the lack of metadata for easier processing in most contents is worse in my opnion.
Two points I want to make clear.
1. Yes, the vast majority of programs are compatible with system-wide screen readers. And for someone with full practical blindness (that is, blindness beyond the point of being able to read even with dark mode or large font), both Windows Narrator and macOS VoiceOver are amazing. Problem is, they read everything. They also fundamentally change the way you interact with the computer, since you now need keyboard shortcuts to control the screen reader and to tell it what to read. If it weren't for the severe astigmatism, I wouldn't need text-to-speech at all...so I certainly don't need it system-wide. I only need it when reading large amounts of text, like when chatting or reading source code.
2. You may prefer the more-true-to-life human voices of things like Google Assistant, and that's okay. But for me, the more important part is being able to read the text at all. If it's more than one or two words then I genuinely can't read it. If I'm trying to troubleshoot my Internet not working, and my cloud-based TTS voice can't talk to its server
because the Internet isn't working, then you can see the problem. I'd prefer my always-working-but-very-robot-y local TTS voice over a true-to-life-but-internet-dependent virtual assistant voice simply because I can always rely on the former to work. The fact that Chrome's TTS voice is cloud-based is a genuine design flaw, not all websites are on the Internet.
What voices do you use when you need to have texts read?
On Windows: Microsoft David Desktop, the default in US/Canada.
On macOS: Daniel, the male British English voice - because it's the voice of those reddit TTS videos and also a classic meme voice.
On Linux: espeak, the built-in male English TTS voice. It's nothing special, and is about as natural-sounding as Stephen Hawking, but it gets the job done. Like I said before, functionality over fidelity.
Programmer's notes:
Windows has a C# API for making Narrator speak arbitrary text. By default, this API uses David Desktop. Both my Visual Studio and Rider plugins use this API and I see no reason to mess with the code to have a different-sounding voice.
Both macOS VoiceOver and espeak have command-line tools for speaking the contents of files, command-line arguments, or the standard input stream. My Rider plugin uses the say command on macOS and the espeak command on Linux, piping a temporary file into standard input which contains the highlighted text I want to be spoken. It'll fill up my /tmp with a bunch of random code snippets, but these get wiped after a reboot and they're generally less than 1KB each if they're larger than a few bytes. I'm not worried.
What other consoles do you have?
Full list:
- Fat PS3 (jailbroken)
- PS5 (disc model)
- Steam Deck (256GB SSD model)
- PS Vita (OLED WiFi model)
- Two slim PS2s, both with fucked disc drives.
- PS1 but it's in storage and I haven't touched it in at least 10 years.
- Two PSPs (3000 model). Both batteries have exploded with age, I don't have an AC adapter, I don't have memory sticks, and one of the PSPs has a cracked screen and the other has a janky zombie UMD drive. They were both CFW'd at one point, though.
Used to have a Kinect model Xbox 360 but we sold it a few years ago. The only game I miss on it is Burnout Revenge, I wish it was on Steam.
I have fond memories of both the PSP and the Vita. PSP was the first console I ever CFW'd and, oronically, the device that executed my first ever line of code at around 9 or 10 years old. It was HTML, sure, but it was both written and executed on a PSP. However, nowadays, both my Vita and PSPs are unplayable...PSPs need repair, and even if they didn't, both the PSP and Vita screens are both too small and not bright enough for me to read text on them.