On-Line Melody Extraction From Polyphonic Audio Using Harmonic Cluster Tracking

Arora, Vipul; Behera, Laxmidhar

doi:10.1109/tasl.2012.2227731

Cited by 23 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the band energy is very simple, it is also a good candidate for enhancement to try features from other melody extraction algorithms. The feature extraction method can also be replaced with others: Hermes' sub-harmonic summation [35], a method based on a log spectrum [36], multi-resolution FFT [37], and so on. Applications of these algorithms for further improvements remain as future work.…”

Section: Resultsmentioning

confidence: 99%

“…In this area of research, challenging tasks exist. These include query-by-singing/humming [3,4], tempo estimation [4,5], cover-song identification [6], music genre classification [7][8][9], music mood classification [10,11], audio fingerprinting [12,13], downbeat estimation [14], audio tagging [15], automatic melody extraction [16][17][18][19][20][21][22][23][24][25], and others. Among the various areas of research in music information retrieval, automatic melody extraction extracts pitch or chroma information from music clips, generally polyphonic music recordings.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic melody extraction algorithm using a convolutional neural network

2017

KSII TIIS

View full text Add to dashboard Cite

In this study, we propose an automatic melody extraction algorithm using deep learning. In this algorithm, feature images, generated using the energy of frequency band, are extracted from polyphonic audio files and a deep learning technique, a convolutional neural network (CNN), is applied on the feature images. In the training data, a short frame of polyphonic music is labeled as a musical note and a classifier based on CNN is learned in order to determine a pitch value of a short frame of audio signal. We want to build a novel structure of melody extraction, thus the proposed algorithm has a simple structure and instead of using various signal processing techniques for melody extraction, we use only a CNN to find a melody from a polyphonic audio. Despite of simple structure, the promising results are obtained in the experiments. Compared with state-of-the-art algorithms, the proposed algorithm did not give the best result, but comparable results were obtained and we believe they could be improved with the appropriate training data. In this paper, melody extraction and the proposed algorithm are introduced first, and the proposed algorithm is then further explained in detail. Finally, we present our experiment and the comparison of results follows.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automatic melody extraction algorithm using a convolutional neural network

2017

KSII TIIS

View full text Add to dashboard Cite

show abstract

“…The identification can be made more robust by considering the temporal evolution of spectra or simply the F0 values [14]. However, for this paper, we consider each frame independently without modeling the temporal dynamics.…”

Section: Identification Of Sourcementioning

confidence: 99%

Instrument identification using PLCA over stretched manifolds

Arora

Behera

2014

2014 Twentieth National Conference on Communications (NCC)

View full text Add to dashboard Cite

Probabilistic latent component analysis (PLCA) is a popular tool for decomposing the spectra of polyphonic music for identifying the constituting musical instruments. Supervised PLCA reconstructs the observed polyphonic spectra using instrument specific spectral parts. However, the distance metrics between two instrument classes in the spectral space may not be the same as that in the acoustic space. This paper proposes a novel geometrical approach to PLCA by modifying the metrics in a way so as to enhance the inter-class distances, while reducing the intra-class distances. Hence, the likelihood function of PLCA is made to give more weight to reconstructing the features which discriminate the sounds of different instruments better. Experiments on binary instrument classification give encouraging results with the proposed approach.

show abstract

“…Existing melody extraction methods can be divided into three categories: source separation-based methods [6,7], data-driven classification-based methods [8,9], and salience-based methods [1,[10][11][12]. Source separation-based methods employ spectrum decomposition schemes to separate the lead voice from the mixed recordings, then estimate and track the pitch sequence of the previously extracted source.…”

Section: Introductionmentioning

confidence: 99%

“…Most melody extraction methods belong to the salience-based category [1,[10][11][12]. Multiple pitches of one music recording are estimated according to some kind of salience function, and tracking strategies are applied to obtain melody pitch sequence taking into account of both salience and smoothness constraints.…”

Section: Introductionmentioning

confidence: 99%

Melody Extraction Using Chroma-Level Note Tracking and Pitch Mapping

2018

View full text Add to dashboard Cite

A new architecture for melody extraction from polyphonic music is explored in this paper. Specifically, chromagrams are first constructed through the harmonic pitch class profile (HPCP) to measure the salience of melody, and chroma-level notes are tracked by dynamic programming. Then, note detection is performed according to chroma-level note differences between adjacent frames. Next, note pitches are coarsely mapped by maximizing the salience of each note, followed by a fine tuning to fit the dynamic variation within each note. Finally, voicing detection is carried out to determine the presence of melody according to the salience of fine-tuned notes. Note level pitch mapping and fine tuning avoids pitch shifting between different octaves or notes within one note duration. Several experiments have been conducted to evaluate the performance of the proposed method. The experimental results show that the proposed method can track the dynamic pitch changing within each note, and performs well at different signal-to-accompaniment ratios. However, its performance for deep vibratos and pitch glides still needs to be improved.

show abstract

On-Line Melody Extraction From Polyphonic Audio Using Harmonic Cluster Tracking

Cited by 23 publications

References 24 publications

Automatic melody extraction algorithm using a convolutional neural network

Automatic melody extraction algorithm using a convolutional neural network

Instrument identification using PLCA over stretched manifolds

Melody Extraction Using Chroma-Level Note Tracking and Pitch Mapping

Contact Info

Product

Resources

About