Comparison of uniform and random sampling for speech and music signals

Zarmehi, Nematollah; Shahsavari, Sina; Marvasti, Farokh

doi:10.1109/sampta.2017.8024356

Cited by 3 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Random sampling is a sampling scheme where sampling positions are randomly taken according to a probability distribution [15,38,56]. It has received a lot of attention in the communities of signal processing, compressive sensing, learning theory and sampling theory, see [13,16,17,21,37,40,42,44,45,59] and references therein. Random sampling of concentrated signals was first discussed by Bass and Gröchenig, and they proved the following result for bandlimited signals concentrated on the cube C R := [−R/2, R/2] d , see [12,Theorem 3.1].…”

Section: Introductionmentioning

confidence: 99%

Random sampling and reconstruction of concentrated signals in a reproducing kernel space

Sun

Xian

2020

Preprint

View full text Add to dashboard Cite

In this paper, we consider (random) sampling of signals concentrated on a bounded Corkscrew domain Ω of a metric measure space, and reconstructing concentrated signals approximately from their (un)corrupted sampling data taken on a sampling set contained in Ω. We establish a weighted stability of bi-Lipschitz type for a (random) sampling scheme on the set of concentrated signals in a reproducing kernel space. The weighted stability of bi-Lipschitz type provides a weak robustness to the sampling scheme, however due to the nonconvexity of the set of concentrated signals, it does not imply the unique signal reconstruction. From (un)corrupted samples taken on a finite sampling set contained in Ω, we propose an algorithm to find approximations to signals concentrated on a bounded Corkscrew domain Ω. Random sampling is a sampling scheme where sampling positions are randomly taken according to a probability distribution. Next we show that, with high probability, signals concentrated on a bounded Corkscrew domain Ω can be reconstructed approximately from their uncorrupted (or randomly corrupted) samples taken at i.i.d. random positions drawn on Ω, provided that the sampling size is at least of the order µ(Ω) ln(µ(Ω)), where µ(Ω) is the measure of the concentrated domain Ω. Finally, we demonstrate the performance of proposed approximations to the original concentrated signal when the sampling procedure is taken either with small density or randomly with large size.

show abstract

Section: Introductionmentioning

confidence: 99%

Random sampling and reconstruction of concentrated signals in a reproducing kernel space

Sun

Xian

2020

Preprint

View full text Add to dashboard Cite

show abstract

Filtering of Noise in Audio/Voice Signal

Thakur¹,

Pandey²,

Gupta³

2018

2018 3rd International Conference on Contemporary Computing and Informatics (IC3I)

View full text Add to dashboard Cite

Speeding up training of automated bird recognizers by data reduction of audio features

Oliveira

Ventura

Ganchev

et al. 2020

PeerJ

View full text Add to dashboard Cite

Automated acoustic recognition of birds is considered an important technology in support of biodiversity monitoring and biodiversity conservation activities. These activities require processing large amounts of soundscape recordings. Typically, recordings are transformed to a number of acoustic features, and a machine learning method is used to build models and recognize the sound events of interest. The main problem is the scalability of data processing, either for developing models or for processing recordings made over long time periods. In those cases, the processing time and resources required might become prohibitive for the average user. To address this problem, we evaluated the applicability of three data reduction methods. These methods were applied to a series of acoustic feature vectors as an additional postprocessing step, which aims to reduce the computational demand during training. The experimental results obtained using Mel-frequency cepstral coefficients (MFCCs) and hidden Markov models (HMMs) support the finding that a reduction in training data by a factor of 10 does not significantly affect the recognition performance. How to cite this article de Oliveira AG, Ventura TM, Ganchev TD, Silva LNS, Marques MêI, Schuchmann K-L. 2020. Speeding up training of automated bird recognizers by data reduction of audio features. PeerJ 8:e8407 http://doi.org/10.7717/peerj.8407 de Oliveira et al. (2020), PeerJ, DOI 10.7717/peerj.8407 2/14Ahmad KS, Thosar AS, Nirmal JH, Pande VS. 2015. A unique approach in text independent speaker recognition using MFCC feature sets and probabilistic neural network.In: Eighth international conference on advances in pattern recognition (ICAPR).

show abstract

Comparison of uniform and random sampling for speech and music signals

Cited by 3 publications

References 11 publications

Random sampling and reconstruction of concentrated signals in a reproducing kernel space

Random sampling and reconstruction of concentrated signals in a reproducing kernel space

Filtering of Noise in Audio/Voice Signal

Speeding up training of automated bird recognizers by data reduction of audio features

Contact Info

Product

Resources

About