<div>Given the powerful influence that music carries in all cultures, it is ideal that everyone have the access and means to learn and understand music. Music visualizations are powerful tools which have proven abilities to help people learn and understand music, as well as enjoy the music. However, when used for educational purposes, it is imperative that the visualization be accurate. This thesis investigated the creation of a visualization which could take input from a live guitar and, in real time, accurately display the pitch within ±2¢, correctly display the note name, and transcribe the note onto the Western music staff. Research was conducted on the history of music visualizations, types of music visualizations and their uses, and commonly used methods of pitch detection.</div><div><br></div><div><div>Processing©3 was selected as the development environment for creating the visualization. Autocorrelation was chosen as the method of pitch detection. Sine waves accurate to 0.01 Hz were generated in MATLAB and used to test the visualization. A tc electronic® PolyTune® 2 guitar tuning pedal was used to tune the guitar before input into the visualization. This served as a means to verify the accuracy of the visualization’s output. A Rocksmith® 1/4-inch to USB cable was used to bring the live guitar signal into the visualization.</div></div><div><br></div><div><div>Processing©3 served as a successful tool for creating the music visualization. The visualization correctly displayed the note name, transcribed the note onto the Western music staff, displayed the audio input, displayed the FFT of the audio input, and accurately displayed the pitch in real time. However, autocorrelation did not give the desired results for pitch accuracy. The detected pitch was not consistently accurate to the desired ±2¢. But, the pitch was accurate to</div><div>at least ±8¢ over the open string range (E2 to E4), and at least ±18¢ over the full range of the guitar (E2 to D6). Though Processing©3 worked well to create the visualization, it may not be the best tool for processing the audio when high accuracy is desired in real time.</div></div>
| Identifer | oai:union.ndltd.org:purdue.edu/oai:figshare.com:article/7498700 |
| Date | 03 January 2019 |
| Creators | Kathryn L. Schmidt (5930840) |
| Source Sets | Purdue University |
| Detected Language | English |
| Type | Text, Thesis |
| Rights | CC BY 4.0 |
| Relation | https://figshare.com/articles/Meaningful_Music_Visualizations/7498700 |
Page generated in 0.0135 seconds