arXiv subjects

Kavya Ranjan Saxena

Publications and source records attributed to Kavya Ranjan Saxena.

3 recordsLinked to original sources

arXiv2025

Uncertainty Quantification in Melody Estimation using Histogram Representation

Confidence estimation can improve the reliability of melody estimation by indicating which predictions are likely incorrect. The existing classification-based approach provides confidence for predicted pitch classes but fails to capture the magnitude of deviation from the ground truth. To address this limitation, we reformulate melody estimation as a regression problem and propose a novel approach to estimate uncertainty directly from the histogram representation of the pitch values, which correlates well with the deviation between the prediction and the ground-truth. We design three methods to model pitch on a continuous support range of histogram, which introduces the challenge of handling the discontinuity of unvoiced from the voiced pitch values. The first two methods address the abrupt discontinuity by mapping the pitch values to a continuous range, while the third adopts a fully Bayesian formulation, which models voicing detection as a classification and voiced pitch estimation as a regression task. Experimental results demonstrate that regression-based formulations yield more reliable uncertainty estimates compared to classification-based approaches in identifying incorrect pitch predictions. Comparing the proposed methods with a state-of-the-art regression model, it is observed that the Bayesian method performs the best at estimating both the melody and its associated uncertainty.

eess.AS↗

arXiv2024

Interactive singing melody extraction based on active adaptation

Extraction of predominant pitch from polyphonic audio is one of the fundamental tasks in the field of music information retrieval and computational musicology. To accomplish this task using machine learning, a large amount of labeled audio data is required to train the model. However, a classical model pre-trained on data from one domain (source), e.g., songs of a particular singer or genre, may not perform comparatively well in extracting melody from other domains (target). The performance of such models can be boosted by adapting the model using very little annotated data from the target domain. In this work, we propose an efficient interactive melody adaptation method. Our method selects the regions in the target audio that require human annotation using a confidence criterion based on normalized true class probability. The annotations are used by the model to adapt itself to the target domain using meta-learning. Our method also provides a novel meta-learning approach that handles class imbalance, i.e., a few representative samples from a few classes are available for adaptation in the target domain. Experimental results show that the proposed method outperforms other adaptive melody extraction baselines. The proposed method is model-agnostic and hence can be applied to other non-adaptive melody extraction models to boost their performance. Also, we released a Hindustani Alankaar and Raga (HAR) dataset containing 523 audio files of about 6.86 hours of duration intended for singing melody extraction tasks.

eess.AS↗

arXiv2023

Deep domain adaptation for polyphonic melody extraction

Extraction of the predominant pitch from polyphonic audio is one of the fundamental tasks in the field of music information retrieval and computational musicology. To accomplish this task using machine learning, a large amount of labeled audio data is required to train the model that predicts the pitch contour. But a classical model pre-trained on data from one domain (source), e.g, songs of a particular singer or genre, may not perform comparatively well in extracting melody from other domains (target). The performance of such models can be boosted by adapting the model using some annotated data in the target domain. In this work, we study various adaptation techniques applied to machine learning models for polyphonic melody extraction. Experimental results show that meta-learning-based adaptation performs better than simple fine-tuning. In addition to this, we find that this method outperforms the existing state-of-the-art non-adaptive polyphonic melody extraction algorithms.

eess.AS↗