Detection of Glottal Closure Instants Based on the Microcanonical Multiscale Formalism

Khanagha, Vahid; Daoudi, Khalid; Yahia, Hussein

doi:10.1109/taslp.2014.2352451

Cited by 30 publications

(30 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The instant of significant excitation (within each period) is termed as Epoch which coincides with instant of closure of the glottis [3]. The problem of detecting the precise locations of such Glottal Closure Instants (GCIs) from speech signal has been studied for decades given its importance in several speech processing tasks [4,5,6,7,8,9,10,11,12]. Most of the successful GCI detectors adopt a two-stage approach -(i) Obtaining an intermediate representation from the speech signal, which explicitly manifests GCIs as discontinuities, impulses, extremas or as other perceptual events and (ii) detecting precise temporal location of glottal closures using custommade algorithms.…”

Section: Background and Previous Workmentioning

confidence: 99%

“…For instance, [5,6,10,13] choose either linear-prediction residual or glottal flow derivative as the representative signal. Other class of algorithms do not explicitly make any model assumption for speech production rather indirectly use the properties of excitation signal (such as its impulsive nature) and estimate appropriate representations (E.g., zero-frequency filtered signal [7], mean-based signal [8], wavelet-decompositions [14], singularity exponents [11]). During the second stage, aforementioned algorithms employ several heuristics to extract (or refine) the GCIs.…”

Section: Background and Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Glottal Closure Instants from Raw Speech Using Convolutional Neural Networks

Goyal

Srivastava²,

Prathosh

2019

Interspeech 2019

View full text Add to dashboard Cite

Glottal Closure Instants (GCIs) correspond to the temporal locations of significant excitation to the vocal tract occurring during the production of voiced speech. GCI detection from speech signals is a well-studied problem given its importance in speech processing. Most of the existing approaches for GCI detection adopt a two-stage approach (i) Transformation of speech signal into a representative signal where GCIs are localized better, (ii) extraction of GCIs using the representative signal obtained in first stage. The former stage is accomplished using signal processing techniques based on the principles of speech production and the latter with heuristic-algorithms such as dynamicprogramming and peak-picking. These methods are thus taskspecific and rely on the methods used for representative signal extraction. However in this paper, we formulate the GCI detection problem from a representation learning perspective where appropriate representation is implicitly learned from the rawspeech data samples. Specifically, GCI detection is cast as a supervised multi-task learning problem solved using a deep convolutional neural network jointly optimizing a classification and regression cost. The learning capability is demonstrated with several experiments on standard datasets. The results compare well with the state-of-the-art algorithms while performing better in the case of presence of real-world non-stationary noise.

show abstract

Section: Background and Previous Workmentioning

confidence: 99%

Section: Background and Previous Workmentioning

confidence: 99%

Detection of Glottal Closure Instants from Raw Speech Using Convolutional Neural Networks

Goyal

Srivastava²,

Prathosh

2019

Interspeech 2019

View full text Add to dashboard Cite

show abstract

“…Then, dynamic-programming or peak-picking is used to select the GCIs among the detected candidates. Examples of such approaches are the SEDREAMS [15], DYPSA [13], DPI [16], YAGA [17], ZFR [14], or MMF [18] algorithms. Although such approaches have been shown to perform reasonably well, they rely on different processing techniques that require manual tuning of parameters (e.g.…”

Section: Introductionmentioning

confidence: 99%

GCI Detection from Raw Speech Using a Fully-Convolutional Network

Ardaillon

Röebel

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Glottal Closure Instants (GCI) detection consists in automatically detecting temporal locations of most significant excitation of the vocal tract from the speech signal. It is used in many speech analysis and processing applications, and various algorithms have been proposed for this purpose. Recently, new approaches using convolutional neural networks have emerged , with encouraging results. Following this trend, we propose a simple approach that performs a regression from the speech waveform to a target signal from which the GCI are easily obtained by peak-picking. However, the ground truth GCI used for training and evaluation are usually extracted from EGG signals, which are not reliable and often not available. To overcome this problem, we propose to train our network on high-quality synthetic speech with perfect ground truth. The performances of the proposed algorithm are compared with three other state-of-the-art approaches using publicly available datasets, and the impact of using controlled synthetic or real speech signals in the training stage is investigated. The experimental results demonstrate that the proposed method obtains similar or better results than other state-of-the-art algorithms and that using large synthetic datasets with many speaker offers better generalization ability than using a smaller database of real speech and EGG signals.

show abstract

“…From the studies in [18], it was observed that most of the epoch detection methods were shown to provide good accuracy on the speech data collected in the lab environments. Also, some attempts were made to see the effectiveness of these methods for additive noise degraded conditions [19][20][21][22][23]. However, there are not many attempts in GCI detection for the degraded conditions like telephone quality speech.…”

Section: Introductionmentioning

confidence: 99%

“…An optimal LoMA is computed within a pitch period using a dynamic programming to locate GCIs. In [19], a nonlinear formalism, namely microcanonical multiscale formalism was used to highlight the impulses present in the speech signal directly.…”

Section: Introductionmentioning

confidence: 99%

A Quantitative Comparison of Epoch Extraction Algorithms for Telephone Speech

Kadiri

2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Telephone speech is one of the degradations involved in building speech systems in practical environments. The potential use of the speech systems depends on the speech analysis algorithms that can handle different acoustic variations and degradations often found in the human speech communication. Detection of epochs/glottal closure instants (GCIs) is typically required in such analysis stages. In this paper, the effect of telephone channel speech on the accuracy of detection of epochs using state-of-art epoch extraction methods is investigated. Epoch is the instant of significant excitation to the vocal tract system in voiced speech. Most of the existing epoch extraction algorithms are shown to perform excellently well on the speech data collected under lab environment. The efficiency of these algorithms for the analysis of telephone quality speech is quantitatively studied and the strengths and weaknesses of the methods are discussed here. The methods are evaluated on six large databases containing speech and simultaneous EGG recordings as the ground truth. The state-of-art epoch extraction algorithms considered in this study for comparison are: ZFF, YAGA, DYPSA, SEDREAMS, SE-VQ and MMF. The performance of the algorithms is evaluated in terms of both reliability and accuracy measures.

show abstract

Detection of Glottal Closure Instants Based on the Microcanonical Multiscale Formalism

Cited by 30 publications

References 32 publications

Detection of Glottal Closure Instants from Raw Speech Using Convolutional Neural Networks

Detection of Glottal Closure Instants from Raw Speech Using Convolutional Neural Networks

GCI Detection from Raw Speech Using a Fully-Convolutional Network

A Quantitative Comparison of Epoch Extraction Algorithms for Telephone Speech

Contact Info

Product

Resources

About