Category: Uncategorized

Well, hello.

AIrchimedes (either Air like in what we breathe, or a melodious A.I.rchimedes like in Eire, in case you were wondering) is a loose collection of pretrained (mostly) GPT2XL (1556) models

• Alan Watts (under the alanwattsbot, or actually, alan_botts) (https://huggingface.co/airchimedes/gpt2xl_alanwatts)
• Arthur Schopenhauer (Parerga and Paralipomena)
• Sokrates

thatcococat is based on the same principles, by the way, and shares a lot with Airchimedes’ approach, but is a bit more of a business idea than an experiment, so to speak

Preparation

Set up a cloud server

generation on K80

Training a model

Get the big serverino

ssh into it

Generation of Text

use free colaboratory Colaboratory GPU Notebooks!

https://colab.research.google.com/drive/1RVwFPLcU9wZMvj3e9Q4VKIVedT6wNP7O#scrollTo=c6ngJZHtYLH7

Today, a little utility for your productivity needs – TomatoTomatoe, a little pomodoro timer for Windows. Check out the pomodoro technique – it’s a nifty lifehack!

The program sits in your system tray and allows you to launch a 20 minute pomodoro, which will count down in a little always-on-top window. At the end, it’ll ask you to describe and rate your pomodoro and write your notes together with the date and start/end time to a csv in a location of your choice (it’ll save your choice). So after a couple of weeks, you’ll be able to import your data into Excel and run the most amazing stats.

Let me know your thoughts! — DOWNLOAD TomatoTomatoe

As i was drifting off to sleep, my thoughts wandered to the topic of eletronically generated instrumental sounds. Think of a violin. Now, think of a soundwave that has (nearly) the same properties as a note played by that instrument. What sound will you “prefer”, i.e. what sound sounds more “natural”?

Do people tend to prefer real instruments over eletronically played versions? Why? What is the difference?

It could be that the subtle difference in the timbre (or the perception thereof) of the sound generated by the resonance of the body of the violin has a different shape than that generated by speakers. Or is the way the shape of the soundwave generated by the violin differs by small yet perceptible details from the eletronically generated one? Has the natural sound of a violin random influences such as the casual vibration of atoms that make up the string, influences that are beyond a trivial imitation by lines of program code and yet noticeable enought? Or is the difference only perceived when the listener know about the different origins of one same sound?

I will investigate this when i wake up.

Here we are.

By going to http://phikappa.eu/jamfm.php you will be promted to save a *.pls file (or it will download automatically, depending on your browser settings).

You can open this pls file in your audio player of choice and it will start streaming the 96 kbps MP3 stream. 🙂

Even if you stop the playback and then get start it again, it should work – however, if you get an 401 error or a password request, just generate a new pls again.

Free the streams – against bloatware. 

Here’s the source code: jamfm.php code as a ZIP file.