Research Article
Transfer Learning for Audio Waveform to Guitar Chord
Spectrograms Using the Convolution Neural Network
Yogesh Jadhav ,
1
Ashish Patel ,
2
Rutvij H. Jhaveri ,
3
and Roshani Raut
4
1
School of Technology Management & Engineering, SVKM’s NMIMS University, Navi Mumbai, India
2
Department of Computer Engineering, Shri Sad Vidya Mandal Institute of Technology (SVMIT), Bharuch, India
3
Department of Computer Science and Engineering, School of Technology, Pandit Deendayal Energy University, Raysan, India
4
Pimpri Chinchwad College of Engineering, Savitribai Phule Pune University, Pune, India
Correspondence should be addressed to Rutvij H. Jhaveri; rutvij.jhaveri@sot.pdpu.ac.in
Received 26 April 2022; Revised 23 June 2022; Accepted 23 July 2022; Published 31 August 2022
Academic Editor: Saqib Hakak
Copyright © 2022 Yogesh Jadhav et al. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Automatic chord recognition has always been approached as a broad music audition task. e desired output is a succession of
time-aligned discrete chord symbols, such as GMaj and Asus2. Automatic music transcription is the process of converting a
musical recording into a human-readable and interpretable representation. When dealing with polyphonic sounds or removing
certain limits, automatic music transcription remains a difficult undertaking. A guitar, for example, presents a greater challenge, as
guitarists can play the same note in a variety of places. e study makes use of CNN functionality to generate the guitar tab;
initially, the constant-Q transform was used to turn the input audio file into short time spectrograms that the CNN model utilises
to analyse the chord. e paper developed a method for extracting chord sequences and notes from audio recordings of solo guitar
performances. For intervals in the supplied audio, the proposed approach outputs chord names and fret-board notes. e model
described here has been refined to achieve an accuracy of 88.7%. e model’s ability to properly tag audio clips is an
incredible advancement.
1. Introduction
e guitar is one of the world’s most popular instruments.
Professional musicians play the guitar, but it is also an
excellent instrument for anybody who enjoys music and
wants to play it freely, even if they have no prior under-
standing of music theory. Although guitars come in a variety
of styles, the bulk of them have six strings. Modern guitar
strings are 65 centimetres in length and tuned to the fol-
lowing pitches: E2 82 Hz, A2 110 Hz, D3 147 Hz,
G3 196 Hz, and E4 330 Hz. e strings of a guitar might
be of nylon or steel, depending on the model. In this work,
we shall use the phrases classical guitar and acoustic guitar
[1, 2] to refer to nylon-stringed guitars that are not electric
guitars.
Each guitar string is slightly unique, not just in diameter
but also in how the strings are produced. On a guitar, the
same pitch can be performed using a variety of string, fret,
and finger combinations. is paper will cover the string/fret
combinations that may be utilised to play the various notes
on classical, acoustic, and electric guitars with six strings.
However, the technique and concepts demonstrated here
may be applied to any other guitar with any number of
strings.
Composing music for the guitar can be done in two ways:
using a classical score or using guitar tablature. Different
amounts of time, fretting, and fingering information are
provided by these two alternatives. For performing a certain
score, there is no one-of-a-kind tablature. A guitar string is
represented by each line in the tablature. e location of the
number indicates the string to pluck, and the number itself
denotes the fret number in the guitar neck where the as-
sociated string must be pushed with a finger of the left hand
to create the required note when plucking the string with the
right hand. It is worth noticing that when the vibrating
component of the string gets shorter, the number increases.
Hindawi
Mobile Information Systems
Volume 2022, Article ID 8544765, 11 pages
https://doi.org/10.1155/2022/8544765