Evolving a Multi-Classifier System for Multi-Pitch Estimation of Piano Music and Beyond: An Application of Cartesian Genetic Programming

Round 1

Reviewer 1 Report

In this manuscript, the authors propose new framework based on a set of classifiers to analyze the audio input and identify the notes present on the given audio signal.

The main advantages of new approach are as follows: It provides simultaneously competitive results even in real-time, white-box optimization and could be applied to other polyphonic instruments.

The manuscript is well-structured.

In the “Introduction” section, the research goal, novelty and contributions are clearly formulated.

The experimental results show the effectiveness of proposed method.

In my opinion, the authors should compare their results with those obtained by using other contemporary audio signals classification algorithms. The selected approaches (Tolonen, 2000; Emiya, 2009; and Klapuri, 2003) in section “6.5. Comparing to other approaches” are outdated.

A comparison with complexity of existing similar classification approaches is also missing.

The implementation details (pseudocode or Python code) are missing.

Future plans are also missing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper presents multi-pitch estimation for music using genetic programming. The use of GP with data augmentation leds to good results.

Overall the paper is interesting. The structure is good and suitable, the methodology is well described and the results are well presented.

My two concerns are:

• The conclusion seem to be just a summary of the whole paper. I suggest to add more like the essential finding, the weaknesses, and future directions.
• The text has various mistakes and flaws of the English language. The most disturbing is the mix between passive voice ("is used", "was designed" etc) and active voice ("we design", "the system uses" etc.)

I marked a few things in the attached PDF, where you should pay attention.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf