New approach to voiced onset detection in speech signal and its application for frame error concealment

Lemyre, C.; Jelinek, Milan; Lefebvre, Rémi

doi:10.1109/icassp.2008.4518720

Cited by 11 publications

(6 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Since the only difference in these two front-ends is the filter bank type, a tentative reason is that the mel triangular filter shape has sharper onset than the gammatone filter, thus may render better characterization of the temporal masking effect. This is based on the theory in [14,15] that human ears tend to focus more on the onset of the power envelope than on the falling edge. We also found (not shown in this work) that in the presence of stationary noise, such as white noise or pink noise, the same experiment had different results: the gammatone+power front-end performed constantly better than the mel+power front-end.…”

Section: Comparison With Other Front-endsmentioning

confidence: 99%

Combined PNCC feature extractor for robust speech recognition

Liu

Zahorian

2014

2014 IEEE China Summit &Amp; International Conference on Signal and Information Processing (ChinaSIP)

View full text Add to dashboard Cite

Recently, two major types of Power-Normalized Cepstral Coefficients (PNCCs) were proposed as noise robust Automatic Speech Recognition (ASR) front-end. All the literatures for these two PNCCs assume clean training data and clean or noisy test data. However, we find that one PNCC method has good performance for the clean training/noisy test scenario, but degrades when test data is cleaner than the training data. The other PNCC method performs relatively better for noisy training/clean test conditions, but is not very robust for the clean training/noisy test conditions. We propose Combined PNCC (C-PNCC) algorithm, which is superior to both previous PNCCs for clean training/noisy test cases, and which also has reasonably good performance for noisy training/clean test conditions.

show abstract

Section: Comparison With Other Front-endsmentioning

confidence: 99%

Combined PNCC feature extractor for robust speech recognition

Liu

Zahorian

2014

2014 IEEE China Summit &Amp; International Conference on Signal and Information Processing (ChinaSIP)

View full text Add to dashboard Cite

show abstract

“…Based on the component energy distributed in specific frequencies, the burst and voicing onsets could be located. The same idea was also utilized by Lemyre et al, 5 who employed a TEO on bandpass filtered signals to derive subband energy profiles. Then the energy profiles were compared to appropriate thresholds to determine the locations of voicing onsets.…”

Section: Introductionmentioning

confidence: 97%

Automatic estimation of voice onset time for word-initial stops by applying random forest to onset detection

Lin

Wang

2011

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

The voice onset time (VOT) of a stop consonant is the interval between its burst onset and voicing onset. Among a variety of research topics on VOT, one that has been studied for years is how VOTs are efficiently measured. Manual annotation is a feasible way, but it becomes a time-consuming task when the corpus size is large. This paper proposes an automatic VOT estimation method based on an onset detection algorithm. At first, a forced alignment is applied to identify the locations of stop consonants. Then a random forest based onset detector searches each stop segment for its burst and voicing onsets to estimate a VOT. The proposed onset detection can detect the onsets in an efficient and accurate manner with only a small amount of training data. The evaluation data extracted from the TIMIT corpus were 2344 words with a word-initial stop. The experimental results showed that 83.4% of the estimations deviate less than 10 ms from their manually labeled values, and 96.5% of the estimations deviate by less than 20 ms. Some factors that influence the proposed estimation method, such as place of articulation, voicing of a stop consonant, and quality of succeeding vowel, were also investigated.

show abstract

“…Предварительная сегментация речевого сигнала на различные фонетические группы используется во многих алгоритмах обработки и кодирования речи [1,2,3,4,5]. Обработка сигнала, учитывающая его характеристики, позволяет улучшить качество звука в устройствах кодирования и декодирования.…”

Section: Introductionunclassified

“…Существует множество алгоритмов разделения речи на различные классы звуков [1,2,3,4,5]. Их общей особенностью является зависимость от характеристик речевого сигнала, для обработки которого они предназначены, и, при применении их к сигналам с отличными характеристиками, качество разделения обычно ухудшается [4].…”

Section: Introductionunclassified

Algorithm for automatic classification of speech segments on based on autocorrelation and energy characteristics

Zhuykov,

Kuznetsov,

Kharchenko

2010

Electron.Commun.

View full text Add to dashboard Cite

The article is devoted to the speech segmentation algorithm by vocal features, based on specifics of autocorrelation function and energy distribution over frequency domain. The algorithm’s classification characteristics are high enough and independent of definite speech base, what demonstrates the proposed algorithm advantage compared to algorithm made for processing of voice with definite characteristics. The operational results with various male and female utterances are considered.

show abstract

New approach to voiced onset detection in speech signal and its application for frame error concealment

Cited by 11 publications

References 2 publications

Combined PNCC feature extractor for robust speech recognition

Combined PNCC feature extractor for robust speech recognition

Automatic estimation of voice onset time for word-initial stops by applying random forest to onset detection

Algorithm for automatic classification of speech segments on based on autocorrelation and energy characteristics

Contact Info

Product

Resources

About