assignments – 18-090, Fall 2019

kabruzzo – Project 1 – Audio Signal Processing With Sensors

kabruzzo@andrew.cmu.edu — Wed, 06 Nov 2019 04:40:38 +0000

For my first individual project, I decided to work with Arduino because I really enjoyed the videos shared on Slack on interactive sound installations. I also have worked with Arduino before, and thought it was cool that you could have it output to Max. I decided to work with flex sensors because I haven’t worked with type of sensor before and thought it was interesting. I also use a potentiometer.

I used a feedback system to create an echo. One of flex sensors pitch-shifts the echo, another one sets the delay of the echo, and the last one sets the volume for the echo. There’s also a degrade that goes from a resolution of 0 bits to 10 bits, then back from 10 bits to 0 bits in a loop. The rate at which the resolution changes is set by the potentiometer.

Video of me changing some of the sensors and the resulting audio

Recording of audio where I used the sensors to process the song Na Na Na by My Chemical Romance

Google drive: https://drive.google.com/drive/u/1/folders/15FI85RAck9Aa7uP0DOmOIj3rCryVGnxx

hfernand – Screaming Ring (Assignment 4)

hfernand@andrew.cmu.edu — Wed, 16 Oct 2019 13:02:58 +0000

For this assignment I attempted to create a ring of spheres that moves faster dependent on the incoming amplitudes from an FFT-ed signal. Using a modified version of the jit.mo organic example, I routed an incoming fft amplitude into the frequency and speed of position attributes for my circles. This means that the frequency and speed of their motion is directly impacted by the amplitude output of the FFT. Ultimately the result is a bit choppy, but I believe this can be attributed to frame dropping.

https://drive.google.com/open?id=1B5iahaCX-J2qdh0D5FXXVcFuQnt2_9Eg

atunnard – assignment 4: playable poly amplitude filter

atunnard@andrew.cmu.edu — Wed, 16 Oct 2019 02:40:12 +0000

this spectral system is a playable, midi-controlled, polyphonic amplitude filter using poly~ and pfft~ in max/msp. each ‘note’ played on the keyboard plays a different slice of the overall amplitude of a sample. the ‘high notes’ are low amplitude sounds of a sample, and the ‘low notes’ are high amplitude sounds. the first sample is an orchestra warming up, looped, and with different amplitude ranges selected with the poly midi keyboard. the second is an excerpt of an old morley grey song, with the same process.

do note: this delightful max tutorial is what allowed me to implement the dual pfft~ greater than/less than amplitude filtration. to this base patch, i added the midi keyboard control and the poly implementation of the amplitude filter, which allows for extremely quirky, layered filtration. the below sounds are cool, but using this patch ‘live’ is the real benefit of it, i’d say.

google drive link

Kabruzzo – Assignment 4: Chlorine fft

kabruzzo@andrew.cmu.edu — Wed, 16 Oct 2019 02:06:11 +0000

For this assignment, I decided to use the cubey-fft patches we made in class as a starting point. Instead of using jit.mo.sin to set the position, I used the amplitudes of the audio signal. I did this by making a matrix with three planes and setting the values in the pfft. I also changed the color and shape generated, and used Chlorine by Twenty One Pilots for the audio signal.

Google drive: https://drive.google.com/drive/u/1/folders/10HVZOsRGqB9YmebxQ9y7pLAmwNyv2J9W

Kabruzzo – Assignment 3: Convolving Drew Gooden

kabruzzo@andrew.cmu.edu — Thu, 03 Oct 2019 06:14:17 +0000

For my original audio, I used the audio from a vine by Drew Gooden. This is the original audio:

My first impulse recording was taken in a hallway in Baker Hall by recording the popping a balloon. The second impulse recording was taken in the courtyard in Mudge using the same method. My first experimental impulse recording was a ringtone. For the second experimental impulse recording I reversed the original audio and added a tempo change of -25%. Here are each of the impulse recordings:

Here are the convolved signals:

google drive link: https://drive.google.com/drive/u/1/folders/1IjURagJg5kqgJC3mnuv69d9FneUeMlrJ

Assignment 3 – Convolve The Girl

hfernand@andrew.cmu.edu — Wed, 02 Oct 2019 12:33:00 +0000

Using the code provided in lecture I convolved a sample of “Kiss The Girl” from The Little Mermaid with four different impulse response samples. The first two impulse responses were recordings of balloons popping in various locations. The first was recorded in a racquetball court and the second was recorded in the Doherty Hall A-Level hallway. The third sample used was a snare drum from a Dubstep sample pack. I chose this sample because I thought its intense and interesting characteristics would be interpreted interestingly with convolution. Finally, I convolved the song with itself to create a horrible reverby noisy mess. The four outputs are linked below, as well as a link to my code and audio files.

https://soundcloud.com/harke-official/p3-convolve-the-girl

https://drive.google.com/open?id=1z_9L-1SVYyMEaXSGvY9rVUS43RD-tb_A

hfernand – Arp Generator (Assignment 2)

hfernand@andrew.cmu.edu — Wed, 18 Sep 2019 07:00:53 +0000

Arpeggiation describes an instrument being played in quick and frequent notes. Using time shifting and the pitch adjustment object from the example patches I’ve created a Max patch that allows you to arpeggiate any sample. It works most effectively with individual notes of an instrument, but can be used to create interesting warping effects with more complicated samples.

Sample used is a free piano loop from Cymatics

Sample and code can be found here: https://drive.google.com/open?id=1GegSX5L9t_hepPwi–p9KOkjNO7kZuGr

atunnard, assignment 2: orchestra warms up: A

atunnard@andrew.cmu.edu — Wed, 18 Sep 2019 04:41:47 +0000

This patch takes an input, runs it into a delay with feedback, then sends the output of that signal into two delays, one in the left channel and one in the right. each of those two signals feeds back into itself, as well as into the input of the opposite side’s signal. for each three of these delay chains, the delay time and pitch modulates randomly. from there, the pitch modulation effect from the feedback is mitigated by a pitch corrector, which fixes each note (or rather, attempts to) to an A. this semi-randomly modulated signal, paired with a nearly-self-oscillating delay chain and enforced pitch effect, creates irregular swells and digital artifacts within a smooth wave of delay.

drive: https://drive.google.com/drive/folders/1fgAeqN-eK4_rT5vyeMRwL-Cu93NiJY-h

sample sound from: https://www.youtube.com/watch?v=RHSbZBuqOvU

Kabruzzo: Assignment 2 – Echos of a Truce

kabruzzo@andrew.cmu.edu — Wed, 18 Sep 2019 04:06:16 +0000

I took the song Truce by Twenty One Pilots and used a delay with feedback. The delay is 0.75s long, and each time it repeats the volume is multiplied by 0.65 and the sampling rate is lowered to 0.4Hz. The resulting effect was similar to an echo. I choose the song Truce because it’s a slower song and I thought it would sound nice with an echo.

Link to google drive: https://drive.google.com/drive/u/1/folders/1TvdzJoAuPSR6QbU16iHvAaTR8X0kJjw-

hfernand – Assignment 1: Do Androids Dream of Electric Bill Gates?

hfernand@andrew.cmu.edu — Sun, 15 Sep 2019 20:52:09 +0000

DeepDream is a type of software created by Google to perform the reverse of facial recognition using a convolutional neural network. Effectively, a neural network trained to recognize faces, animals, or object takes an image and adjusts it slightly to “enhance” what it perceives to be patterns resembling the data it is trained on. A free piece of software that performs this task is available at deepdreamgenerator.com, and this is what I used for my found system. I used an image of the Gates Hillman Center from the Carnegie Mellon Press Kit, and ran it through the Deep Dream Generator 25 times. The Deep Dream generator has multiple options for neural networks that you can use, and different images that you can “train” the network on. I opted to use different network settings for different loops, as I didn’t want the final image to consist of only one type of pattern. This resulted in a highly warped image consisting of what appear to be dogs, fruits, and colorful spirals. In the video below I showcase each iteration of the generator’s output, and list the network setting I used.

The collection of produced images can be found here: https://drive.google.com/drive/folders/1gBPbAUWQjrRO948-1sF7Zf4l6LjDVhHp?usp=sharing