We overhauled our website to include all of our research. Sub pages were used to organize the information by topic, and we hope the website is easy to navigate.
Our presentation today went fantastic. The time we spent rehearsing paid off, and I hope we were able to illustrate our topic as much as we could without boring our audience. It was exciting to see the collective effort of our project come together for our presentation.
We met today and further looked into experiments with varying sampling rates and bit depths of recordings. Using Audacity, we exported files at different bit rates and sampling frequencies, as well as in different formats. The result was VERY obvious. Once the sampling rate dropped below 20K, the recording sounded more and more like it was played through an AM radio, chopping off the top end of the recording. Lowering the bit rate resulted in a very techno version of the sound that also sounded muffled, as if underwater. We drew up graphs to illustrate both sampling rates and bit depth, and added more to our Google Slide. We will meet tomorrow to finish the slides, as well as rehearse in the presentation room.
We added information to our Google Slide and will be adding it to the website shortly.
I cleaned up the presentation slides from the last meeting with the group. Further researched differences between .MP3 files and .WAV files. I also looked into the history of sampling rates. 44.1K, 48K, and 96K are common standards of music industry. 44.1K is the result of human hearing being limited to around 20K Hz, and since both ends of the wave needs to be recorded, doubling it (and adding some head room) resulted in 44.1K being the standard sampling frequency.
A change in plan:
We decided that looking at the individual bits of an audio file was a bit incomprehensible. While it would have been nice to look directly at an audio file, even an .mp3 file (which is a tenth the size of a raw audio file) is huge.
We decided to look at how audio works in a computer at a more manageable scale. We viewed a video on YouTube that talked about the audio engineering of sound. It wasn't exactly out topic, but it had two major points that we will need to address, which is bit depth, and sampling frequency.
We researched more into how audio is stored from a microphone. What a microphone does is convert sound waves in the air to an electrical signal by use of a diaphragm. This electrical signal is recorded by a computer sound card, and turned into the numbers that are stored in a computer ( the same numbers we were looking at earlier.) These numbers are a measure of the voltage every some thousand times a second. The number of times a computer records the voltage is the sampling frequency.
We looked into the most popular forms of audio files, which are .WAV, .MP3, and Flac files. Each one differs mostly in their size. An .MP3 is a lossy compression file, which reduces the size of the file by eliminating inaudible sounds in the file. A Flac is a lossless compression file, which retains the information, but only compresses the file by about half the size. .WAV is the standard uncompressed file for a Windows PC.
We made an outline for our presentation slides, and will expand on it hopefully later in the week.
Today, we made our website. It is not complete, but we plan to edit it as we go further into our research. We started with opening .mp3 files with notepad as a .txt file, and found that it was composed of ASCII symbols.
We experimented by copying parts of the data back onto itself, as well as deleting some. The results always gave a jarring noise, followed by jumping to different parts of the song. An odd side effect was that the parts of the song that we left intact had strange noise added.
We will meet again soon to look into why the audio file is changing the way it is.