Linear-scale filterbank for deep neural network-based voice activity detection

Jung, Yongju; Kim, Younggwan; Lim, Hyungjun; Kim, Hoirin

doi:10.1109/icsda.2017.8384446

Cited by 12 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the success of MFCCs for both speech and non-speech applications, some studies suggest that GTCCs outperform MFCCs in non-speech audio recognition (Valero and Alias, 2012) and in applications featuring diverse acoustic environments (Bonet-Solà and Alsina-Pagès, 2021). LFCCs have also been shown to consistently outperform MFCCs in applications that contain higher-frequency information of interest (Jung et al, 2017; Lei and Lopez-Gonzalo, 2009; Zhou et al, 2011).…”

Section: Methodsmentioning

confidence: 99%

An unsupervised data-driven approach for wind turbine blade damage detection under passive acoustics-based excitation

Solimine

İnalpolat

2022

Wind Engineering

View full text Add to dashboard Cite

Existing passive acoustics-based techniques for wind turbine blade damage detection lack the robustness and adaptability necessary for an operational implementation due to their physics- and model-based dependency. In contrast, this study develops an entirely unsupervised, data-driven damage detection technique. The novelty of the technique lies in (i) the development and comparison of spectral and cepstral-domain features for the robust characterization of the cavity-internal acoustics, (ii) the use of autoencoder networks to reduce the effects of non-stationary acoustic excitation, and (iii) the exclusion of labeled or damage-case data in the training set. The technique was successfully demonstrated on a wind turbine blade section inflicted with damage of various sizes, types, and locations, and subjected to airflow-induced passive acoustic excitation provided by a wind tunnel. Damage detection accuracy up to 99.82% was achieved for some damage types.

show abstract

Section: Methodsmentioning

confidence: 99%

An unsupervised data-driven approach for wind turbine blade damage detection under passive acoustics-based excitation

Solimine

İnalpolat

2022

Wind Engineering

View full text Add to dashboard Cite

show abstract

“…The LLFB features are extracted by a classical pipeline for filterbank-based features. Specifically, a signal goes through a pre-emphasis filter; is segmented into (overlapping) frames and a Hamming window function is applied to each frame (the frame length is 25 ms and the frame step size is 10 ms); afterwards, we use a 256 points short-time Fourier Transform (STFT) on each frame and calculate the power spectrum; and subsequently compute and apply a linear-scale filter banks with 80 triangular overlapping windows where center frequencies of the windows are equally spaced along a Hz scale [13]. By taking the logarithm of the power spectrogram and truncating these utterance-level features with the same processing used for the CQT feature, LLFB features with unified shapes of 80 × 400 (where 400 is the number of time frames) can be obtained.…”

Section: Front-end Processingmentioning

confidence: 99%

Multi-task Learning-Based Spoofing-Robust Automatic Speaker Verification System

Zhao

Togneri

Sreeram

2022

Circuits Syst Signal Process

View full text Add to dashboard Cite

Spoofing attacks posed by generating artificial speech can severely degrade the performance of a speaker verification system. Recently, many anti-spoofing countermeasures have been proposed for detecting varying types of attacks from synthetic speech to replay presentations. While there are numerous effective defenses reported on standalone anti-spoofing solutions, the integration for speaker verification and spoofing detection systems has obvious benefits. In this paper, we propose a spoofing-robust automatic speaker verification system for diverse attacks based on a multi-task learning architecture. This deep learning-based model is jointly trained with time-frequency representations from utterances to provide recognition decisions for both tasks simultaneously. Compared with other state-of-the-art systems on the ASVspoof 2017 and 2019 corpora, a substantial improvement of the combined system under different spoofing conditions can be obtained.

show abstract

“…Since the vocal information of infant cry is distinctly characterized in the time-frequency domain, such as melodies and formants [14], [15], we considered a spectrogram-based representation of audio signals as the input of CNNs for feature learning. In this study, the log linear-scale filterbank (LLFB) spectrogram [16] was used to represent audio signals.…”

Section: Introductionmentioning

confidence: 99%

Convolutional Neural Networks for Audio-Based Continuous Infant Cry Monitoring at Home

et al. 2021

View full text Add to dashboard Cite

published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User

show abstract

Linear-scale filterbank for deep neural network-based voice activity detection

Cited by 12 publications

References 13 publications

An unsupervised data-driven approach for wind turbine blade damage detection under passive acoustics-based excitation

An unsupervised data-driven approach for wind turbine blade damage detection under passive acoustics-based excitation

Multi-task Learning-Based Spoofing-Robust Automatic Speaker Verification System

Convolutional Neural Networks for Audio-Based Continuous Infant Cry Monitoring at Home

Contact Info

Product

Resources

About