Ai

(in case you just want the github link without reading the hallucinations of my neural network - here you go) I hate voice memos. Mostly I hate to receive them. You can’t quickly glance over to remember the content of the conversation, the key piece of information you need is hidden in a 3minute voice note. But I recently learned that talking to machines is quite practical. I even tweeted (its forever gonna be tweeting) about talking to an LLM while pooping. Not that I haven’t used the feature before a few times but I never got the hang of it because I like to brain dump but then I want to read the reply (my brain works funny). I’ve even created a bot a while back that would transcribe signal voice memos that I braindumped to it, but it didn’t stick, Until I’ve come across hyprwhspr about a week or two ago. I immediately liked the idea, I’ve done several whisper projects in the past, whisper-rs-cli just this week so I dug into it. I wanted it as lean as possible so I first created a stripped down fork of it and fitted it to my fedora/ubuntu setup. But it was somewhat slowish in python and I’ve just came out of doing whisper-rs-cli so the decision was obvious. I need a rust version. Given that I had a working python version that I liked I thought its gonna be a slightly more straightforward path. It wasn’t. Specially when I went down the path of “lets add real time typing into any app just for giggles” path. Since I’ve done some real(ish) time transcription api experiments before I thought that will also just be an extra half an hour. ...

Today I came across this tweet about speed reading and I was immediately intrigued. I was a fan of it back in my younger days when I wanted to consume as much knowledge as possible. I never had the patience to become very proficient in it but I did pick up a thing or two. I’m still generally a fairly fast reader, even when I read for pleasure. But who wouldn’t want to read more books? ...

I’ve created a subpage on this site dedicated to various tools (the subset of them that are mature enough for me to dare mention them) -> you can find them here. I do not commit to any level of regular maintenance of them, they get fixed when they need fixing (which is usually when I’m in a rush to do something and I realize everything is broken). Occasionally these things find a life of their own, I’ve created a simple script to have data retention policies in free version of mattermost in January 2018 for a project I was consulting on and promptly forgot about it. Over the years it appeared that several people found it rather useful - it got regular updates, support for additional database type and more. That is a long way of saying feel free to submit a PR.

A quick tutorial for people who want to run goose but want it to keep your data confidential you generally have two options. First its to buy a good gpu to run a model locally, the second one is use advancement in trusted execution environments and use confidential AI backend with it. How to do it Go to PrivateMode and create an account and an API key. Then, set the environment variable PRIVATE_MODE_API_KEY to your API key. You can do this in your terminal with the following command: ...

I spend a considerable amount of my time in terminal(s) and I’ve gotten used to incorporating large language models into my workflow. But occasionally its just too annoying to switch over to a browser to ask it something. My previous solution for that was that I had a telegram bot that I’d chat with (which was convenient and faster than web) but I felt something was missing. And today with my procrastination spiking I’ve finally solved it! ...

In a world of explosive amount of content about AI and where NOT having an AI newsletter has become a rarity there are a few authors and sites I like to read for their more thoughtful and long form content: Recommended Resources The following sites are highlighted for substantive coverage: gwern.net - Features deep learning content alongside other topics ym2132.github.io - Focuses on extended posts about generative adversarial networks simonwillison.net - Combines Python and AI topics retrochronic.com - Explores intersections of AI and philosophy Video Content For visual learners, the author recommends the AI Explained YouTube channel as an additional resource. ...