1. Team 17: Learning Musical Instruments through Spectral and Temporal Analysis

2. (Best Project) No.

3. (Top 3) No.

4. (Clarity, Organization) 6

5. (Project Idea) 10

6. (Research Contribution) 2

	When I started reading your report, I was excited to see you were working on autonomous differentiation of musical instruments. From the way you wrote the paper, I would guess you are a sound engineer or perhaps you perform in an orchestra? The first few pages are strong and full of content. Past those pages the report was not as good.

	What appears to have happened as you were working on this project is that you start ambitiously and with a brilliant idea (hence the 10/10 for project idea), but because you run out of time and resources you end up with some preliminary results and an 18 page paper (14 pages of which are either huge, full-page diagrams or references). An example of this is the diagram which almost fills all of page 2. The diagram is not of statistics or results, it is a diagram containing 13 large images of instruments. This would not be acceptable in a research publication.

	I do not know how proficient you were with sound imaging programs before you started the project, but if you are familiar with them you know that you could probably create the diagrams which occupy the bulk of your paper in two or three hours. In the report you say you used Matlab to obtain the graphs. Matlab has all sorts of built-in ways to accomplish that task.

	Going solo on the project is not much of an excuse for inadequate work. Many of the other students did their work on their own and created some of the top projects in the class.

	I do not see anywhere in the paper any evidence that you actually implemented a learning algorithm. I don't see much more than a visualization of the sound and some formulas to explain how you "will" process the sound. It would have been a much nicer project if you had completed the plan laid out in Figure 4 (Instrument Recognition Diagram).

	This line explains the whole paper: "These features will be used to determine the class of instrument that the sound came from." The key word there being "will". This is written more like a proposal than a report, and seeing as it is mostly composed of sound visualization images it would be a weak proposal.

	The grammar is good but not perfect, and the content definitely needs work. Next time, try to start work on the project much sooner and start getting results at least a week before you need them so as to have plenty of time to work on the report.

	Given the briefness of your report and the lack of results, I don't have much more to say about your project. Score: 3/10