I worked with a blind dev and his screen reader is incomprehensible by everyone but him. It had to be at like 5-6x. Guy was a wizard. Sure he had some disadvantages with vision but the guy was wicked fast and would pump out the most code

As someone who also does this, I find that the problem becomes not processing the information, but retaining it, and I haven't found a solution to that yet.

I doubt regular folks can reach 5x or even 2x with good retention. Perhaps if it's a podcast of people chatting about random things, it's possible, but any intellectually stimulating content you probably will have problems. Visually impaired people probably have an advantage on account of having their entire visual cortex freed up and possibly able to help with this (not sure if it does, neuroscience majors here?)

(As a sighted person) I watch all my videos or podcasts at 2x. Depending on how new or dense the content is, I might have to occasionally dial it back, but I think I do fairly well at 2x (usually, this is only the case if I'm distracted or the slides are going by too quickly as well). Often I can even up it to 3x.

Maybe retention is worse, but I'm still better off. Higher speeds are more engaging and enjoyable to me, as bad speakers and long pauses matter less. Therefore I benefit from watching more talks and lectures. For better retention it is preferable to occasionally re-watch good talks for spaced repetition.

The Video Speed Controller extension is indispensable part of my browsing experience, mostly for keyboard shortcuts to speed up/down and << >> videos.

I used to listen/watch everything at 1.5x-2x speed but found a similar result. Now I only increase the speed if it's information I only need to process once (e.g the news) but not remember later (lectures, tutorials etc).

I've been doing that lately (1.5x-2x), but taking notes, and making sure to work on any exercises provided after the lecture. I need notes for retention even at 1x, so it doesn't feel like extra work.

The problem with retaining is independent of processing information speed.

I've heard screen readers going way faster (I don't know quantitatively) than what it sounds like when I listen to videos at 2x.

I also think it's important that the screen readers use a "old school" synthesized voice, instead of a voice that sound natural, because each of the phonemes is articulated distinctly.

I do that, too.

Especially presentations. For most movies, I am only comfortable with 1.25x. In presentations people talk extra slowly. It is a classic advice when learning to give presentations, talk slower. After getting used to 2X speed, it becomes extremely frustrating to listen to a live presentation.

Although even 2x presentations still feels slower than my reading speed as kid. But I cannot speed read anymore since I need glasses

I've had the same experience with audio books. I find it's easy with entertainment. In normal speech there is a lot of redundancy. You can miss entire sentences and still follow. I don't this would be the case with a screen reader.

You got me curious.

I just put some news interview on YouTube at 2x.

It is fun how even the accents are easily recognizable.

Anyway, this is in Spanish. My guess by reading the comments is Spanish is a much easier language to listen at a faster speed.

If you want to try it, older tts voices such as eloquence are intelligible at much higher speeds than the modern natural sounding voices.

Personal attacks will get you banned here, so please don't do that, even if another comment rubs you the wrong way.

https://news.ycombinator.com/newsguidelines.html

Asking for personal decency toward another human being is a banning offense here?

"Thanks for taking the time to share how superhuman you are" is not "asking for personal decency".

Please don't post in the flamewar style to HN. We're here for curious conversation, which doesn't go along with the style of argument in which people cast each other's comments in the worst possible light. If you review the site guidelines you'll see that many of them guard against that argument style. That's no accident, because it's so common in online discussion generally, indeed has become the default. It takes conscious work to have a place that doesn't fall into it. That's what we're trying for.

Note this one, for example: "Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith." If you'd review https://news.ycombinator.com/newsguidelines.html and take the intended spirit to heart, we'd be grateful.

The person took the time to mention their ability to listen and comprehend two times faster than presumably the average. They didn't not mention the original poster in any sense of compassion, ignoring them.

What is the strongest possible interpretation of what this person said? That which is not a fair understanding of the criticism of my post?

The plausible interpretation is that the commenter got excited about screen readers because they unexpectedly connected to something he was already interested in, and it made him curious. Not only is that a perfectly decent thing, it's an example of the curious conversation that HN exists for.

Meanwhile, snarking and putting down other users, which unfortunately are what you did, are examples of what HN very much does not exist for.

It's really easy to interpret other people's internet comments in ways that add things that weren't necessarily there. We have only tiny blobs of text to go by. We don't know each other, and we don't even have the cues that come from voice, body language, and the like. As I read the GP comment, it seems to me that it was you who introduced the notions that it was somehow about the commenter puffing himself up, or that he was somehow lacking in compassion towards the original poster.

In a thread of 200 comments, there will inevitably be rivulets of conversation that aren't about the main theme. Such tangents are only bad if they're somehow generic and predictable. This one was interesting, still very much in the orbit of the main topic, and I really don't think there was any meanness in it. I'm pretty trained to pattern-match meanness after years of trying to get people to be kind to each other on HN, and I'm afraid it was your comment that was the mean one.

Without meaning it as a personal attack, do you have any reason to think that 6 years of trying to detect meanness has made you any more accurate at it than you were before, or than anyone else is?

When you say "trained to pattern match it", is there any feedback to change your models? As moderator-by-fiat it's your decision what comments are mean, which is potentially a self reinforcing loop.

Here's some feedback: under Dang's eye, HN has become a paragon of online discussion. I say this as someone who has occasionally been called out.

He didn't say that he can listen two times faster than the average person. He said that he can listen to podcasts at twice their recorded speed. There is nothing superhuman about this. Many people do so. People naturally speak slower than their listeners can understand, and practiced public speakers are trained to speak even slower than that.

Listening to speech at double speed is really not that noteworthy, as anyone can do it with a bit of practice. I didn't construe it as bragging, just as outlining the thought process which led them to think that it might be worth them trying a screen reader themselves.