davidcoo@andrew.cmu.edu – 18-090, Fall 2019

Davidcoo – Project 2: Leap Motion Granular Synth

davidcoo@andrew.cmu.edu — Mon, 09 Dec 2019 06:23:41 +0000

In my project I created a granular synthesizer that is controlled using the leap motion. I started by using a base of the Sugar Synth that we used in class and built on that to be able to control it with using leap motion controls. By utilizing the built in detection of the leap motion, the user is able to control which value they are controlling by doing gestures with their hands (Open hand, fist and the OK symbol), Then they use their other hand to control the value by moving left and right to assign a value to the parameter they have chosen. There is a toggle to pick which hand that is controlling the value and which is controlling the parameter. The parameters that can be controlled are the speed, pitch rate, random pitch, grain size, and number of grains.

Drive Link: https://drive.google.com/drive/u/1/folders/1Zz5Ca0h4jl3yKpqmj89UUT3WMRukfpj2

In my project I used the Sugar Synth: http://formantbros.jp/sako/download.html

I also used the leap motion for max and built on the leap motion made in class.

Davidcoo – Project 1: Super Convolution

davidcoo@andrew.cmu.edu — Wed, 06 Nov 2019 09:08:18 +0000

My project preforms video convolution that takes in a video to be convoluted and the convolution is based on. The “convolution video” (matrix the convolution is based on) is calculated by taking the average of the RGB values and converting them to a float that is in 0 to 1.

It can be scaled so that the videos can be broken up into multiple pieces that convolve independently of each other. This is done through the poly~ object. In the poly~, I take sub matrixes of each of the videos that correspond to the portion of the video we want to view and convolve them.

In the sample video I provided I convolved the built in basketball video with the built in roller blading video. In my patch I scale the resolution of the video to 80×60. This is because the patch uses math that involves a lot of matrices in real time which is very intensive I spent so much time waiting and having Max crash on me . I also included a patch that will scale to any size video but most videos are at a resolution that is too intense so it probably won’t work (but the math checks out). To test the realistic version use super-conv and the project1-main patches.

The sample demonstrates the patch working in 2×2, 2×1, and 1×1 “convolution resolution”.

https://drive.google.com/drive/u/0/folders/1xjNnMdg0R6cHRm8qa02h7HfTatXLNZmL

Davidcoo – Assignment 4 – Chimes

davidcoo@andrew.cmu.edu — Wed, 16 Oct 2019 06:41:10 +0000

In this project I used pfft~ to create an output for a buffer~ object. This works similarly to how we did time shifting. The bin number indexes over the buffer~ object and reads in order to convert its amplitudes into cartesian form. By doing this we can create sound from the buffer.

In the buffer I periodically added an impulse to a random index in the buffer at random index in a variable range. Then at longer intervals I randomly scaled down all of the amplitudes by using “apply gain”.

Then I took this signal and ran it through another pfft~. This pfft~ is a band pass filter for amplitude. First I compared the amplitude of the signal to the low value of the filter and let it pass if it was higher. Then I compared to the high value and let the signal pass if it was lower.

In my patch, when setting the filter to a lower range you can hear more frequencies but at a lower volume. When the filter contain a range containing high amplitudes you can hear less frequencies but louder. Setting the range for the impulse signals higher creates signals that seem out of my hearing range, and a lower range sounds more pleasing.

Link to Drive: https://drive.google.com/drive/u/0/folders/1AKL4V9agJPkGc5qryCKyd9aKcqO0Lqbw

The first pfft~ was inspired by this youtube video. I changed the behavior of how the scaling of the amplitudes works. I also changed the timings so more frequencies can be heard. https://www.youtube.com/watch?v=qkT2il6UiYw

Davidcoo – Assignment 3 – It’s Time to Convolve it

davidcoo@andrew.cmu.edu — Wed, 02 Oct 2019 03:40:05 +0000

In this projected we used an impulse signal to convolve a piece of audio. This creates a sound that resembles what it would sound like if it happened where the impulse single was recorded.

The first impulse response I have was recorded in Porter Hall near the entrance on Frew St. Here the ceiling is curved creating a stretched out dome. There is also a hallway on either side that has a similar ceiling which creates a long echoing effect. It was created by recording the sound of a balloon popping. Then I normalized the sound and passed it into the max patch we were provided with.

The second impulse signal was recorded using the same method between the Resnik and West Wing buildings. At this location there are curved wall on each side which create an echoing effect.

The the third and fourth impulse responses are not actual impulse responses. The third “Impulse Response” is a police siren. I thought this would sound more interesting but it just sounds like a siren that is echoing.

The fourth is a recording of thunder I found online. This caused the voice in the convoluted recording to sound booming as if a higher being is talking to you.

Sources: https://www.youtube.com/watch?v=9Elem7z8yVw
https://www.youtube.com/watch?v=T-BOPr7NXME

Davidcoo – Project 2: Speech Jammer

davidcoo@andrew.cmu.edu — Wed, 18 Sep 2019 06:12:32 +0000

This project was inspired by when I watch a video about a speech jammer on youtube. A speech jammer works by playing back audio from a microphone to the speaker at a small delay, causing the speaker to stumble in their talking. An online version of this can be found here. https://www.clicktorelease.com/code/speech-jammer/

To make this project I first started out by playing back audio at a slightly delayed time. Then I created a video that would also play back at a delay. I used math to sync up these delays. In order to make it more difficult to speak when using this system, I also created a feedback for the audio and an echo for the video. I connected both of these to scale appropriately.

When the delay is around 100 ms it is hardest to keep speaking normally, so the frame delay should be set to 32. However, if you want to experiment with the delay you can change it to any number from 0-32 where 0 is a longer delay and 32 is a shorter one.

I would encourage you to give it a try! Note that it is a lot harder to talk normally at first but you can get used to it pretty quickly. Also speaking off the top of your head is harder than doing something like reading. Have fun!

https://drive.google.com/drive/folders/1U9A_EXxT6iOHkBCyfod56D1LfGcDc8D9

Davidcoo – Assignment 1: Cup String Phone

davidcoo@andrew.cmu.edu — Tue, 03 Sep 2019 21:33:24 +0000

The system I chose for this assignment was creating a string telephone. To make this I used two cups and some string. I poked holes in the bottom of each cup and passed the string through it. Then I tied knots in the string so it would not come out of the cup. For the string telephone to work the cups have to be far enough apart so that the string is tight.

For the original messaged I talked directly into the cup and had my friend record the audio on his phone from the other side. To make more generations I had him play the audio back on his phone speaker into the cup while I recorded the other cup with my phone. We did this until there was 9 generations and the audio is completely unrecognizable.

Although most of the distortion was because of the cup and string system, a good deal of it was also done due to the phones’ microphones and their speakers.