Human speech may have a universal transmission rate: 39 bits per second • Hacker News Board

Re: Human speech may have a universal transmission rate: 39 bits per second

#151

Post by samstave » Thu, Sep 05, 2019, 1:38 AM UTC

sweetbacon wrote: post #139

dredmorbius wrote: post #91

Earlier quoted context omitted.

Exceedingly slow narration, particularly what's normal for audiobooks, is annoying to me because it's slower than I process words. It's like walking with someone whose pace is far slower than your natural gait -- it takes more energy and concentration to slow down. It's why slow-talkers are so annoying. This isn't "how fast can I go through this" but "what is a comfortable pace"? So I bump the speed up, though usuall…

As a slow walker I respectfully disagree on the energy to slow down vs speed up! Concentration may be equal however. I get absolutely exhausted keeping up with you fast walkers. :)

Found the slow walker!

But let me give you an annecdote:

Ive buiilt a lot of buildings BIG buildings (ten hospitals, data centers and HQ offices for facebook google brocade lucas salesforce etc...)

I am a fucking FAST walker. Because i need to get shit done bitch!

You have a million square feet on 49 stories? Get the fuck up! When you have 400,000 square feet and you need to show thos LV or AV contractors where to be and youre working on a schedule - speed the fuck up!!

Yeah i dont do well with slow walkers.

Re: Human speech may have a universal transmission rate: 39 bits per second

#152

Post by Merrill » Thu, Sep 05, 2019, 1:38 AM UTC

Typing at 60 words/minute, 5 characters/word and 8 bits/character gives a gross bit rate of 40 bits/second.

Of course, the information rate is a lot less, since there are fewer than 8 bits of information per character in English. The paper says "from 4.8 bits per syllable for Basque to 8.0 bits per syllable for Vietnamese" and there are multiple characters per syllable. So the typing information bit rate is probably somewhere around 10 to 15 bits/second.

Re: Human speech may have a universal transmission rate: 39 bits per second

#153

Post by thrwayxyz » Thu, Sep 05, 2019, 1:38 AM UTC

ilamont wrote: post #15

But, he says, instead of being limited by how quickly we can process information by listening, we’re likely limited by how quickly we can gather our thoughts. That’s because, he says, the average person can listen to audio recordings sped up to about 120%—and still have no problems with comprehension. Some years ago I worked on an accessibility project for an app and website designed for people with disabilities. One…

I'm fully sighted but I use espeak tts at around 1000wpd for fluff text like ecnomist articles and 300wpm for heavy going text like the text sections of math books.

I also watch most video at 2/3 times the speed since the skills seem transferable.

Re: Human speech may have a universal transmission rate: 39 bits per second

#154

Post by ahicks » Thu, Sep 05, 2019, 1:46 AM UTC

ilamont wrote: post #15

But, he says, instead of being limited by how quickly we can process information by listening, we’re likely limited by how quickly we can gather our thoughts. That’s because, he says, the average person can listen to audio recordings sped up to about 120%—and still have no problems with comprehension. Some years ago I worked on an accessibility project for an app and website designed for people with disabilities. One…

I'm the blind dev who refactored a huge chunk of the Rust compiler [0]. I'm at roughly 800 words a minute with a synth, with the proven ability to top out at 1219. 800 or so is the norm among programmers. In order to get it we normally end up using older synths which sound way less natural because modern synthesis techniques can't go that fast. There's a trade-off between natural sounding and 500+ words a minute, and the market now strongly prefers the former because hardware can now support i.e. concatenative synthesis.

1219 is a record as far as I know. We measured it explicitly by getting the screen reader to read a passage and dividing. I spent months working up from 800 to do it and lost the skill once I stopped (there was a marked level of decreased comprehension post 1000, but I was able to program there; still, in the end, not worth it). When I try to described the required mental state it comes out very much like I'm on drugs. Most of us who reach 800 or so stay there, though not always that fast for i.e. pleasure reading (I do novels at about 400). it's built up slowly over time, either more or less explicitly. I did it because I was in high school doing muds and got tired of not being able to keep up; it took about 6-8 months of committing to turn the synth faster once a week no matter what, keeping it there and dealing with a day or two of mild headaches. Note that for most blind people these days, total synthesis time per day is around 10+ hours; this stuff replaces the pencil, the novel, etc. Others just seem to naturally do it. You have little choice, it's effectively a 1 dimensional interface, so from time to time you find a reason to bump the knob. And that's enough.

Whether and how much the skill transfers to normal human speech, or even between synths, is person-specific. I can't do Youtube at much beyond 2x. Others can. It's definitely a learned skill.

0: https://ahicks.io/posts/April%202017/rust-struct-field-reord...

Re: Human speech may have a universal transmission rate: 39 bits per second

#155

Post by aidenn0 » Thu, Sep 05, 2019, 1:46 AM UTC

Merrill wrote: post #152

Typing at 60 words/minute, 5 characters/word and 8 bits/character gives a gross bit rate of 40 bits/second. Of course, the information rate is a lot less, since there are fewer than 8 bits of information per character in English. The paper says "from 4.8 bits per syllable for Basque to 8.0 bits per syllable for Vietnamese" and there are multiple characters per syllable. So the typing information bit rate is probably…

IIRC 1.3 bits per character is the value used as a rule-of-thumb for english text entropy, which would yield 6.5bps at 60WPM and an average stenographer can go at 4x that rate for approximately 26bps.

Re: Human speech may have a universal transmission rate: 39 bits per second

#156

Post by drdeadringer » Thu, Sep 05, 2019, 1:52 AM UTC

This has at-least tangential interest to me as one of my side projects is to create a custom "Number Station". Maybe out of a Raspberry Pi or similar, I'm still very much in the preliminaries. Since I'm in the US I am aware of the FCC.

Re: Human speech may have a universal transmission rate: 39 bits per second

#157

Post by ahicks » Thu, Sep 05, 2019, 1:58 AM UTC

ahicks wrote: post #154

ilamont wrote: post #15

But, he says, instead of being limited by how quickly we can process information by listening, we’re likely limited by how quickly we can gather our thoughts. That’s because, he says, the average person can listen to audio recordings sped up to about 120%—and still have no problems with comprehension. Some years ago I worked on an accessibility project for an app and website designed for people with disabilities. One…

I'm the blind dev who refactored a huge chunk of the Rust compiler [0]. I'm at roughly 800 words a minute with a synth, with the proven ability to top out at 1219. 800 or so is the norm among programmers. In order to get it we normally end up using older synths which sound way less natural because modern synthesis techniques can't go that fast. There's a trade-off between natural sounding and 500+ words a minute, and…

And as a followup to that--because really this is the weird part--some circles of blind people (including mine) talk faster between ourselves. That's not common, but it happens. I still sometimes have to remember that other people can't digest technical content at the rate I say it and remember to slow down. A good way to bring it out is to have me try to explain a technical concept that I understand really well. I have the common problem in that situation of not being able to talk as fast as I think, but I also seem to have the ability to assemble words faster in a sort of tokenize/send to vocal cords sense once I know what I want to say.

To me, the fact that this does in fact seem to be bidirectional at least some is more interesting than that I can listen fast.

Re: Human speech may have a universal transmission rate: 39 bits per second

#158

Post by ashes-of-sol » Thu, Sep 05, 2019, 1:59 AM UTC

samstave wrote: post #151

sweetbacon wrote: post #139

Earlier quoted context omitted.

As a slow walker I respectfully disagree on the energy to slow down vs speed up! Concentration may be equal however. I get absolutely exhausted keeping up with you fast walkers. :)

Found the slow walker! But let me give you an annecdote: Ive buiilt a lot of buildings BIG buildings (ten hospitals, data centers and HQ offices for facebook google brocade lucas salesforce etc...) I am a fucking FAST walker. Because i need to get shit done bitch! You have a million square feet on 49 stories? Get the fuck up! When you have 400,000 square feet and you need to show thos LV or AV contractors where to be…

I don't do well with people who don't realize I'm walking slow sure to my spastic muscle disorder giving me a bad day. You need to dial your attitude down several notches.

Re: Human speech may have a universal transmission rate: 39 bits per second

#159

Post by devnulloverflow » Thu, Sep 05, 2019, 2:01 AM UTC

ilamont wrote: post #15

But, he says, instead of being limited by how quickly we can process information by listening, we’re likely limited by how quickly we can gather our thoughts. That’s because, he says, the average person can listen to audio recordings sped up to about 120%—and still have no problems with comprehension. Some years ago I worked on an accessibility project for an app and website designed for people with disabilities. One…

Well I'm sure the ability requires training, but I wonder if it is specific to screen readers.

Consider what you quote: > we’re likely limited by how quickly we can gather our thoughts

Now the amount of relevant info on a screen is typically small enough that a sighted person can zero in on it at a glance and perhaps just click a button without thinking.

I.e. the amount of info that deserves "gathering our thoughts" is typically very small. So if that is the bottle-neck, your colleague can keep cranking up the audio speed until low-level processing audio becomes the bottle-neck, which is a regime that sighted people never deal with even, not even the nerds who speed up their Joe Rogan podcasts.

Re: Human speech may have a universal transmission rate: 39 bits per second

#160

Post by taneq » Thu, Sep 05, 2019, 2:01 AM UTC

ahicks wrote: post #154

ilamont wrote: post #15

But, he says, instead of being limited by how quickly we can process information by listening, we’re likely limited by how quickly we can gather our thoughts. That’s because, he says, the average person can listen to audio recordings sped up to about 120%—and still have no problems with comprehension. Some years ago I worked on an accessibility project for an app and website designed for people with disabilities. One…

I'm the blind dev who refactored a huge chunk of the Rust compiler [0]. I'm at roughly 800 words a minute with a synth, with the proven ability to top out at 1219. 800 or so is the norm among programmers. In order to get it we normally end up using older synths which sound way less natural because modern synthesis techniques can't go that fast. There's a trade-off between natural sounding and 500+ words a minute, and…

Wow, that's incredible. Do you find it frustrating talking to actual humans now? I'd imagine it feels like they're speaking in slow motion.

Edit: Hah, just saw your post on talking faster to other people who have the same audio skills.