greg76

Audio sleep stories

• Llama-3.2-3B-Instruct-4bit is used for generating the story based on the instructions in PROMPT.md
  • Llama 3.2 seems to work best with smaller stories: keeping the stories around 1000 words results ~4 min audio, with a slightly slower speed cadence.
• Kokoro-82M-bf16 is used as the text to speech model (using even the "premium" built-in macOS voices produced weaker results with the say command)
• Add your own sound file (eg. BACKGROUND_SOUND = "rain.m4a") for background soundscape.
• There were some weird type innotation bugs in some of the underlying packages, hence the # type: ignore comments for the linter.

greg76

from abc import ABC, abstractmethod
from dataclasses import asdict, dataclass


class Character(ABC):
    @dataclass(slots=True, kw_only=True)
    class Stats:
        STR: int
        DEX: int
        CON: int

greg76

EXCERPT: Signal//Noise

by [Your Name]

The dome of the Grand Assembly cracked with strobes and pulses of pure algorithmic light. Hexsound tunneled through the crowd like a virus—compressed, modulated, fractal-burst into flesh. In the center pit, on a circular dais of jury-rigged silicon and scrapmetal chrome, the Sceners were at war.

"Sectorjam’s up," said Ryn, thumb twitching on her neural-ink interface. She sat in the mezzanine shadows above the compo pit, visor lenses flickering with debug streams. “Running custom synthmods on a gutted Triarc Z80. He’s bending opcode like origami.”

Below, Sectorjam—neon mohawk, skin like a failed BIOS boot—drove his fingers across a ghostboard, each tap spitting machinecode into a corrupted 8-bit memory space. His demo was alive: vector forests blooming in real-time, synced to audio generated from nothing but electromagnetic ripple and clever abuse of voltage.

“Watch the sync chain,” muttered Kaz, Ryn’s partner, hunched beside her, his spinal jack trailing

greg76

;
; assembly example to write a message on the Commodore 64 screen
;

; address of the basic loader. the RUN command is looking for the first instruction here.
; the assembly code should be compiled to start here as well, so it can be run with a sys command
*=$0801

; this is the basic loader, this is mimicing of jumping in to the assembly code with a sys command
!byte $0c,$08,$b5,$07,$9e,$20,$32,$30,$36,$32,$00,$00,$00

greg76

"""\
Boids is an artificial life program, which simulates the flocking behaviour of birds, and related group motion.

Some resources I'd recommend if you wanted to learn more about boids:
* https://en.wikipedia.org/wiki/Boids
* https://people.ece.cornell.edu/land/courses/ece4760/labs/s2021/Boids/Boids.html
* http://www.red3d.com/cwr/boids/

This is my quick excercise to make a simplified, 2d, retro inspiried implementation.
"""

greg76

import os
import time
import random

animals = "🐅🐆🦓🦣🐘🦛🦏🐪🦒🦘🦬🐃🐂🐄🐎🐖🐏🦙🐐🦌🐕🐈🐓🐇🦝"
#animals = "🐶🐱🐭🐰🦊🐻🐼🦁🐯🐨🐻🐮"

width = os.get_terminal_size().columns - 2
pos = [width] * len(animals)

greg76

#!/usr/bin/env bash
#
# Bootstrap script for setting up a macOS machine
#

# Requires xcode and tools!
xcode-select -p || exit "XCode must be installed! (use the app store)"


echo "Starting bootstrapping"

greg76

import os
import datetime

with os.popen("nvram -p") as out:
    ts_line = next((line for line in out.readlines() if line.startswith("panicmedic-timestamps")), None)
    if ts_line:
        ts_hex = ts_line.split("\t")[-1]
        ts_int = int(ts_hex, 16)
        ts = datetime.datetime.fromtimestamp(ts_int // 1000000)
        print(f"Last kernel panic: {ts}")

greg76

import collections

with open("words.txt") as f:
  # let's read the words from the file, but ignore anything that ends with s
  # just to remove plurals
  words = [line.strip() for line in f.read().splitlines() if line[4] != "s"]

  # how often the different letter occur?
  occurances = collections.Counter()
  for word in words:

greg76

import locale

import requests
import tabulate

locale.setlocale(locale.LC_ALL, "en_US")

# available snapshots: 30, 90, 365 days
data_ranges = {"new": 30, "old": 90}