Noise robust audio surveillance using reduced spectrogram image feature and one-against-all SVM

Sharan, Roneel V.; Moir, T. J.

doi:10.1016/j.neucom.2015.02.001

Cited by 24 publications

(22 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While time and frequency domain features continue to be used in ASR systems, such as in audio surveillance applications [8,17], they are often only used as supplementary features.…”

Section: Time and Frequency Domain Featuresmentioning

confidence: 99%

“…The number of filters in the bank is normally in the range of 20-24 with 19 filters used in [2], 20 filters in [8,20], 23 filters in [10,17] and 24 filters in [9]. In addition, the first and second derivatives of the coefficients, commonly known as delta and delta-delta coefficients, respectively, which provide trajectories of MFCCs over time, are often appended to improve the classification performance of ASR systems.…”

Section: Mfccsmentioning

confidence: 99%

“…In the event γ ¼ 1, the filter bank energies are uncompressed which is often referred as linear cepstral coefficients. When evaluated on an audio surveillance database in [8], linear MFCCs were shown to give significant improvement in classification accuracy at low signal-to-noise ratios (SNRs) and also found to be more effective for feature vector combination.…”

Section: Mfccsmentioning

confidence: 99%

“…The SIF with reduced feature dimensions, referred as reduced SIF (RSIF), is proposed for an audio surveillance application in [8]. For the RSIF, the mean and standard deviation values of the central moment values are computed along the rows and columns of the blocks.…”

Section: Central Momentsmentioning

confidence: 99%

“…However, applications have diversified since then with interests in areas such as audio surveillance [8], sound event recognition [9], and environmental sound recognition [10]. While audio surveillance can be seen as a subclass of sound event recognition, there are some differences as discussed in Section 4.1.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

An overview of applications and advancements in automatic sound recognition

Sharan

Moir

2016

Neurocomputing

Self Cite

View full text Add to dashboard Cite

a b s t r a c tAutomatic sound recognition (ASR) has attracted increased and wide ranging interests in recent years. In this paper, we carry out a review of some important contributions in ASR techniques, mainly over the last one and a half decades. Similar to speech recognition systems, the robustness of an ASR system largely depends on the choice of feature(s) and classifier(s). We take a wider perspective in providing an overview of the features and classifiers used in ASR systems starting from early works in content-based audio classification to more recent developments in applications such as sound event recognition, audio surveillance, and environmental sound recognition. We also review techniques that have been utilized in noise robust sound recognition systems and feature optimization methods. Finally, some of the less commonly known applications of ASR are discussed.

show abstract

“…While time and frequency domain features continue to be used in ASR systems, such as in audio surveillance applications [8,17], they are often only used as supplementary features.…”

Section: Time and Frequency Domain Featuresmentioning

confidence: 99%