Music Genre Classification Using Transfer Learning

Liang, Beici; Gu, Minwei

doi:10.1109/mipr49039.2020.00085

Cited by 22 publications

(8 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the outcome of tonal analysis one can distinguish not only the key signature corresponding to the examined piece of music [8], [22], [34], but also the graphical representation of the content associated with that piece [6], [35], or various graphical imaging of its chords [36]. It is worth mentioning that in recent years we have had an opportunity to witness the rapidly growing interest in data analysis solutions applying all sorts of machine learning techniques [37]- [39], including methods based on artificial neural networks [21], [40]. Such a tendency has also been observed in the area of music data analysis oriented towards musical genre recognition [28], [37], [39]- [41].…”

Section: Related Workmentioning

confidence: 99%

“…It is worth mentioning that in recent years we have had an opportunity to witness the rapidly growing interest in data analysis solutions applying all sorts of machine learning techniques [37]- [39], including methods based on artificial neural networks [21], [40]. Such a tendency has also been observed in the area of music data analysis oriented towards musical genre recognition [28], [37], [39]- [41].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Trajectory of Fifths in Music Data Mining

Kania

Łukaszewicz

2021

IEEE Access

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Trajectory of Fifths in Music Data Mining

Kania

Łukaszewicz

2021

IEEE Access

View full text Add to dashboard Cite

“…In a work by Beici Liang and Minwei Gu [2], a similar problem of audio-based classification of 11 western music genres, including Rock, Pop, Rap, Country, Folk, Metal, Jazz, Blues, R&B, Electronic Music and Classical Music was taken up and was done using a transfer learning approach. An advantage of this approach is that the pre-trained models that were obtained from the source task of music auto-tagging can be successfully adapted to the target task in order to achieve high performance measurements.…”

Section: Related Workmentioning

confidence: 99%

Music Genre Classification using Optimized Sequential Neural Network

2021

IJATCSE

View full text Add to dashboard Cite

Music makes up a huge portion of the contents stored and used over the internet, with several sites and applications developed solely to provide music-related services to their users/ customers.Some of the most challenging tasks in this scenario would include music classification based on languages and genres, playlist suggestions based on music history, song suggestions based on playlist contents, top genres / songs based on listeners' rating, likes, number of streams, song loops, popularity of artists based on number of songs released per year, hit songs per year, etc. One of the most important stages to solve the above-mentioned challenges would be music genre classification. It would be impractical to analyze each and every song in a given database to identify and classify music genres, even though human beings are better at performing such tasks. Hence, useful Machine Learning algorithms and Deep Learning approaches may be used for accomplishing such tasks with ease. A thorough analysis to understand the different uses of Machine Learning and Deep Learning algorithms and relevance of such algorithms with respect to situations would be made to highlight and contrast the advantages and disadvantages of each approach. The outcomes of the optimized models would be visualized and comparedto the expected outcomes for better perception.

show abstract

“…Finally, they observed an AUC value of 0.894 given by an ensemble classifier that combined both approaches. In [5], authors B. Liang and M. Gu have tackled the problem of audio-based classification using a transfer learning approach. They evaluated multiple models on a dataset of 1100 songs with 11 genres.…”

Section: Literature Reviewmentioning

confidence: 99%

Using Machine Learning to Classify Music Genre

Nihalaani¹

2021

IJRASET

View full text Add to dashboard Cite

As Plato once rightfully said, ‘Music gives a soul to the universe, wings to the mind, flight to the imagination and life to everything.’ Music has always been an important art form, and more so in today’s science-driven world. Music genre classification paves the way for other applications such as music recommender models. Several approaches could be used to classify music genres. In this literature, we aimed to build a machine learning model to classify the genre of an input audio file using 8 machine learning algorithms and determine which algorithm is the best suitable for genre classification. We have obtained an accuracy of 91% using the XGBoost algorithm. Keywords: Machine Learning, Music Genre Classification, Decision Trees, K Nearest Neighbours, Logistic regression, Naïve Bayes, Neural Networks, Random Forest, Support Vector Machine, XGBoost

show abstract

Music Genre Classification Using Transfer Learning

Cited by 22 publications

References 1 publication

Trajectory of Fifths in Music Data Mining

Trajectory of Fifths in Music Data Mining

Music Genre Classification using Optimized Sequential Neural Network

Using Machine Learning to Classify Music Genre

Contact Info

Product

Resources

About