History of Speech Spectrum Analysis

Fulop, Sean A.

doi:10.1007/978-3-642-17478-0_3

Cited by 10 publications

(14 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The limitations of the short-time power spectrum that stem from the well known trade-off between time and frequency resolution have been extensively discussed in the literature [8], [20]. Time-frequency reassignment addresses this timefrequency trade-off, offering an improved representation of the temporal evolution of spectral components.…”

Section: Coefficientsmentioning

confidence: 99%

“…Among these various time-frequency distributions, the timefrequency reassignment is a method that improves the representation of the speech spectral content, as it is very useful in representing simultaneously the temporal, i.e.,, onsets of plosive sounds, and the spectral features, i.e.,, harmonic structure of vowels, of speech signals [8]. In addition, when the recognized speech signal is impinged by reverberation, its spectral envelope, and therefore the MFCC features that describe this envelope, are smoothed and carry less information.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A reassigned front-end for speech recognition

Tryfou

Omologo

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Abstract-This paper introduces the use of the TFRCC features, a time-frequency reassigned feature set, as a front-end for speech recognition. Compared to the power spectrogram, the time-frequency reassigned version is particularly helpful in describing simultaneously the temporal and spectral features of speech signals, as it offers an improved visualization of the various components. This powerful attribute is exploited from the cepstral reassigned features, which are incorporated in a state-ofthe-art speech recognizer. Experimental activities investigate the proposed features in various scenarios, starting from recognition of close-talk signals and gradually increasing the complexity of the task. The results prove the superiority of these features compared to a MFCC baseline.

show abstract

Section: Coefficientsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A reassigned front-end for speech recognition

Tryfou

Omologo

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…Moreover, in AAC the shape of the transform window can be adaptively selected between a sine window and a Kaiser-Bessel-derived (KBD) window (Fulop, 2011):…”

Section: Lossy Compressionmentioning

confidence: 99%

Detection of Montage in Lossy Compressed Digital Audio Recordings

Korycki¹

2015

Archives of Acoustics

View full text Add to dashboard Cite

This paper addresses the problem of tampering detection and discusses methods used for authenticity analysis of digital audio recordings. Presented approach is based on frame offset measurement in audio files compressed and decoded by using perceptual audio coding algorithms which employ modified discrete cosine transform. The minimum values of total number of active MDCT coefficients occur for frame shifts equal to multiplications of applied window length. Any modification of audio file, including cutting off or pasting a part of audio recording causes a disturbance within this regularity. In this study the algorithm based on checking frame offset previously described in the literature is expanded by using each of four types of analysis windows commonly applied in the majority of MDCT based encoders. To enhance the robustness of the method additional histogram analysis is performed by detecting the presence of small value spectral components. Moreover, computation of maximum values of nonzero spectral coefficients is employed, which creates a gating function for the results obtained based on previous algorithm. This solution radically minimizes a number of false detections of forgeries. The influence of compression algorithms' parameters on detection of forgeries is presented by applying AAC and Ogg Vorbis encoders as examples. The effectiveness of tampering detection algorithms proposed in this paper is tested on a predefined music database and compared graphically using ROC-like curves.

show abstract

“…where is the time lag and represents the complex conjugate of the signal [43]. In essence, the Winger-Ville distribution is the Fourier transform of the instantaneous autocorrelation.…”

Section: Wigner-ville Distributionmentioning

confidence: 99%

“…Haar, Daubechies, Coiflet, Symlet, Meyer, Morlet and Mexican Hat are some wavelet family instances of mother wavelets. Thus, the appropriate mother wavelet must be chosen according to the particular application [43].…”

Section: Three-level Wavelet Decomposition Treementioning

confidence: 99%

Patient-specific epileptic seizure detection in long-term EEG recording in paediatric patients with intractable seizures

Zabihi

Kıranyaz

İnce

et al. 2013

IET Intelligent Signal Processing Conference 2013 (ISP 2013)

View full text Add to dashboard Cite

Over recent years, due to the increase in the epileptic patient population, issues of diagnosing and treatment of epilepsy have become more and more prominent and much research has been done in this field in consequence. However, there are still many gaps and lack of knowledge in interpreting Electroencephalograph (EEG) signals in order to solve the problem.Particular problems in this area include difficulties in detecting the seizure events (due to the different seizure types and their variability from patient to patient or even in an individual over time), and dealing with long-term EEG recordings, which is an onerous and time consuming task for electroencephalographers.The thesis discusses the two problem areas using EEG data from four subjects with overall 21 hours of recording from the CHB-MIT scalp benchmark EEG dataset. We propose a patient specific seizure detection technique, which selects the optimal feature subsets, and train a dedicated classifier for each patient in order to maximize the classification performance. To exploit the characteristics of a patient's EEG pattern as much as possible, we used a large set of features in the proposed framework, namely time domain, frequency domain, time-frequency domain and nonlinear features, and selected the most crucial features among them by using Conditional Mutual Information Maximization (CMIM) technique. We further performed extensive comparative evaluations against 6 other feature selection methods to demonstrate the superiority of the CMIM.Support Vector Machine (SVM) with the linear kernel is used as the classifier. The experimental results show a delicate classification performance over the test set, i.e. an average of 90.62% sensitivity and 99.32% specificity are acquired when all channels and recordings are used to form a composite feature vector. In addition, an average sensitivity and specificity rates of 93.78% and 99.05% are obtained using CMIM, respectively. PREFACEThis master thesis was carried out in the MUVIS group at the signal processing department of Tampere University of Technology. It was funded by the Academy of Finland.I wish to express my sincere gratitude to Professor Serkan Kiranyaz, whose inspiring guidance and inexhaustible patience have helped me cultivate a systematic approach towards the research. I am indebted to him for the freedom he gave me in implementing my ideas. I have also acquired valuable insights through his instruction in academic approach.I am very much grateful to Professor Turker Ince for his warm support and his valuable suggestions. Besides, I would also like to thank Professor Moncef Gabbouj for providing me with the opportunity for the thesis work. He gave me a lot of trust and flexibility in the project.I have been extremely fortunate in having valuable friends who have made a homely atmosphere for me during my stay in Tampere

show abstract

History of Speech Spectrum Analysis

Cited by 10 publications

References 31 publications

A reassigned front-end for speech recognition

A reassigned front-end for speech recognition

Detection of Montage in Lossy Compressed Digital Audio Recordings

Patient-specific epileptic seizure detection in long-term EEG recording in paediatric patients with intractable seizures

Contact Info

Product

Resources

About