cpmpercussion

% LaTeX Template for short student reports.
% Citations should be in bibtex format and go in references.bib
\documentclass[a4paper, 11pt]{article}
\usepackage[top=3cm, bottom=3cm, left = 2cm, right = 2cm]{geometry} 
\geometry{a4paper} 
\usepackage[utf8]{inputenc}
\usepackage{textcomp}
\usepackage{graphicx} 
\usepackage{amsmath,amssymb}  
\usepackage{bm}  

cpmpercussion

#!/bin/bash
# This one-liner script plays a video in an infinite loop on a raspberry pi
# for a video-art exhibition.
# It was written for a video in portrait orientation so the video is rotated 270
# to use up the whole screen (which was also rotated).
# Charles Martin, July 2016

omxplayer -o local --loop /home/pi/video.mp4 --orientation 270

cpmpercussion

#!/usr/bin/env python3
"""FAIS Minus --- a script for handling final marks exported from Wattle Gradebook

OVERVIEW:

This script helps process the final mark output from Gradebook, including:

- generating stats per cohort 
- recording and updating special grades (e.g., DA/RP/WD)
- exporting CSV files that are _actually_ SAS-ready.

cpmpercussion

\documentclass[a4paper,12pt,oldfontcommands]{memoir}
\PassOptionsToPackage{hyphens}{url}
\usepackage{url}
\usepackage{hyperref}
\usepackage{doi}
\usepackage{graphicx}
\usepackage{amssymb}
\usepackage{epstopdf}
\usepackage{wrapfig}
\usepackage{natbib}

cpmpercussion

from music21 import converter, instrument, note, chord, stream, midi
import numpy as np
import pandas as pd
import os
import time
import h5py
import argparse
import multiprocessing

parser = argparse.ArgumentParser()

cpmpercussion

Putting Pd on the Intel Galileo

The Intel Galileo is a great little development/experiments board with an Intel Quark 32-bit x86 processor. We had some of these hanging around our office and I was inspired to make some computer music with it - so here's my process for getting Pure Data up and running.

Here's the things I needed:

• Intel Galileo Gen 2
• MicroSD card
• $2 Ebay usb audio interface (1 in, 2 out)
• Sparkfun FTDI basic adapter

cpmpercussion

#!/bin/bash
echo "$1"
videoname=${1%%.*}
#echo "$videoname"
ffmpeg -i $1 -c:v libx264 -b:v 2M -minrate 0.5M -preset fast -vf scale=1920:1080 -c:a aac -b:a 128K ${videoname}_1080.mp4
ffmpeg -i $1 -c:v libx264 -b:v 0.5M -minrate 0.5M -preset fast -vf scale=960:540 -c:a aac -b:a 128K ${videoname}_540.mp4
ffmpeg -i $1 -y -codec:a libmp3lame -ac 2 -ar 44100 -ab 64k -id3v2_version 3 -metadata title="$videoname" -metadata album_artist="Charles Martin" -metadata album="COMP2300 2020" -metadata genre="Vocal" -metadata date="2020" ${videoname}_audio.mp3
echo "done, phew."

cpmpercussion

Datasets

• Harmonix Set - Beats, downbeats, and functional structural annotations for 912 Pop tracks.
• da-tacos - a dataset for cover song identification and understanding

Tools

• deezer/spleeter - Spleeter is the Deezer source separation library with pretrained models written in Python and uses Tensorflow

cpmpercussion

from tensorflow.python.client import device_lib

def get_available_gpus():
    local_device_protos = device_lib.list_local_devices()
    return [x.name for x in local_device_protos if x.device_type == 'GPU']

get_available_gpus()

cpmpercussion

DOCNAME=nime-paper-input.md
OUTNAME=rendered-output
REFSFILE=references.bib
ARGS=--bibliography $(REFSFILE)

nime:
	pandoc --template=nime.latex --self-contained --natbib --filter pandoc-fignos $(DOCNAME) -o $(OUTNAME).tex
	sed -i '.bak' 's/\citep{/\cite{/g' $(OUTNAME).tex
	pdflatex $(OUTNAME)
	bibtex $(OUTNAME)