Step 0: You will be making Korean stake. Step 1: Mix those ingredients. Step 2: Now that you mixed those ingredients, do something else.
System started doing Step 1, believed it was over so moved to Step 2 and when was asked to go back, kept going back to step 2.
Step 1 being Step 0 and Step 1 combined also works.
Again, it's also a weird way to prerecord. If you're prerecording, you're prerecording all steps and practicing with them prerecorded. I can't imagine anyone to be able to go through a single rehearsal with prerecorded audio to not figure out how to do this, we have the technology.
There's a simple explanation that isn't 'prerecorded'. I'd be very happy to accuse Meta of faking a demo, but that's 1) just a weird way to fake a demo and 2) effect that has easier explanation.
You ask AI how to do something. AI generates steps to do that thing. It has concept of steps, so that when you go 'back' it goes back to the last step. As you ask how to do something, it finishes explaining general idea and goes to first step. You interrupt it. It assumes it went through the first step and won't let you go back.
The first step here was mixing some sauces. That's it. It's a dumb way to make a tool, but if I wanted to make one that will work for a demo, I'd do that. Have you ever tried any voice thing to guide you through something? Convincing Gemini that something it described didn't happen takes a direct explanation of 'X didn't happen' and doesn't work perfectly.
It still didn't work, it absolutely wasn't wi-fi issue and lmao, technology of the future in $2T company, it just doesn't seem rigged.
Thanks for checking it out. It could definitely use larger library of sounds in general.
Ideally, I'd like to allow sharing and storing of presets, but it was simply out of scope for the PoC - the functionality is there in the desktop version btw, but it on the other hand asks users to download an unknown .exe and then share mp3 and json files with each other, putting us firmly in the mid-90s'.
I come to shill my webapp for background noise because it has a twist not appearing in anyone's recommendations.
Whenever I'm switching between tasks (thinking vs reading vs writing) I'd either turn the sound off or on, given I needed more or less attention at the moment. Minor problem with that was that sometimes unexpectedly I'd stick with the new task longer than expected, start to get bored, but w/e background sound I had on didn't match the task, so I'd look for something else... Overall a bit annoying for some groups of tasks.
I'm experimenting with mixing music with podcasts with extra noise and turning it on and off, but I also made https://stimulantnoi.se/ (with extra reading on psychological basis of the design and link to open source standalone desktop app on https://incentiveassemblage.substack.com/p/why-is-nobody-ser...). It allows for mixing (including uploading additional) sounds into sets and binds switching between those whole sets to media keys for quick access.
I have no idea how did I miss them last time I was looking around, unless they grew significantly over last half a year or so. I'll check it out when I get back to this project, thanks.
I wish I was hiring, if that's what you're asking ;) Otherwise, if you have any ideas for processing formulas (even just for reading them out, but any extra steps towards expressing what they mean - ' 'sum divided by count' is 'mean'/'average' value ' being the most simple example I can think of) I'd love to hear them. Novel ideas in technical papers are often expressed with formulas which aren't that complicated conceptually, but are critical to understanding the whole paper and that was another piece I was having very mixed results with.
My use case is research papers. That means very clear text, combined with graphs of varying form and quality and finally occasional formulas.
Two approaches I had most, but not full, success with are:
1) converting to image with pdf2image, then reading with pytesseract
2) throwing whole pdfs into pypdf
3) experimental multimodal models
You can get more if you make content more predictable (if you know this part is going to be pure text just put it in pypdf, if you know this is going to be a math formula explain the field to the model and have it read it back for high accessibility needs audience) the better it will go, but it continues to be a nightmare and a bottleneck.
Time of day (or time after waking up per subject) when tests were administered has not been controlled. Cognitive abilities are mediated by wakefulness (not to mention, related for most people, digestive processes) cycle.
If '"Night owls" smarter than morning people' sounds more plausible than 'time since waking up and last meal predictive of cognitive performance' it's time to get one's identity checked. And I can't imagine 'journalists' from thrash like Sky (Guardian this time) not knowing that, which brings me to the final point: what is this link doing here?
Time of day (or time after waking up per subject) when tests were administered has not been controlled. Cognitive abilities are mediated by wakefulness (not to mention, related for most people, digestive processes) cycle.
If '"Night owls" smarter than morning people' sounds more plausible than 'time since waking up and last meal predictive of cognitive performance' it's time to get one's identity checked. And I can't imagine 'journalists' from thrash like Sky not knowing that, which brings me to the final point: what is this link doing here?
https://en.wikipedia.org/wiki/AI_effect We got it named already, it just needs to be properly propagated until there's no value left in calling things 'AI'.
Author mentions, but doesn't focus on, 'work being too challenging/not challenging enough'. I wrote a fair bit about it here (with a slightly different focus as name suggests, but I go over original research first) https://incentiveassemblage.substack.com/p/why-is-nobody-ser.... I'm not sure why 'challenge level' is less focused on compared to lack of interruptions - both seem about equally demanding to environment including manages and take similar amount of work to adjust.
Either way, to save you a click, Csikszenthmihalyi research wasn't mainly about cognitive load, because we already had a fair bit of research on cognitive load. It seems insufficient (although I do have my reservations), but addition of complexity of the task and w/e additional issues are happening is pretty solid predictor* of performance. Challenge/skill 'graph' presented can be reinterpreted with challenge/skill on X axis and a parallel flat line above it. Even better, and empirically supported, graph can be seen in first image in the post I linked, but it is a bit much to paint with words.
Flow research is cool, but there are more simple and actionable tools.
*Observant reader may notice that this is because of lack of units, but we do have physiological indicators if one desires to monitor them.
For reference, I'm transcribing research-related podcasts, meaning speech doesn't overlap a lot, which would be a problem for WhisperX from what I understand. There's also a lot of accents, which are straining on Whisper (though it's also doing well), but surely help WhisperX. It did have issues with figuring out the number of speakers on it's own, but that wasn't a problem for my use case.
With all seriousness, if we give up on solving the problem we would like to see, and try to solve what decision makers see as their problem, providing feed management tools (also?) to managers probably would be an improvement.
I've been working on white noise app for a bit, but with a slightly different focus. While it's maintaining the same core points (limited bullshit, hopefully clear interface, sound staying on with phone screen being off) it addresses problem I constantly found myself having with every app: if I want to switch anything, I have to go back to the webpage/app and there goes some of my focus. Link: https://stimulantnoi.se/
So I made an app focused (hehe.) on ease of switching between noise intensity. There's a long explanation why it's important if you want one (https://incentiveassemblage.substack.com/p/why-is-nobody-ser...), but if you're the kind of nerd who knows about 'flow state' and 'Yerkes-Dodson law', you pretty much got everything covered. The core point is: if intensity of your main activity is changing, your background noise should too, so that you maintain the same total level of arousal. Most likely double so for ADD people.
My current solution is to use media buttons (forward/backward) to control intensity. It took much more work than you'd expect to get media interface to do this without breaking currently playing sounds. In general anything that isn't 'Play this sound' works much worse than one would anticipate with how prevalent media is on the internet - I see ambiphone does the same thing with playing sounds separately, but you saved yourself from managing media interface, so I can't quite tell how much pain exactly you have experienced with this.
System started doing Step 1, believed it was over so moved to Step 2 and when was asked to go back, kept going back to step 2.
Step 1 being Step 0 and Step 1 combined also works.
Again, it's also a weird way to prerecord. If you're prerecording, you're prerecording all steps and practicing with them prerecorded. I can't imagine anyone to be able to go through a single rehearsal with prerecorded audio to not figure out how to do this, we have the technology.