Analysis of Acoustic Events in Speech Signals Using Bessel Series Expansion

In this paper, we propose an approach for the analysis and detection of acoustic events in speech signals using the Bessel series expansion. The acoustic events analyzed are the voice onset time (VOT) and the glottal closure instants (GCIs). The hypothesis is that the Bessel functions with their damped sinusoid-like basis functions are better suited for representing the speech signals than the sinusoidal basis functions used in the conventional Fourier representation. The speech signal is band-pass filtered by choosing the appropriate range of Bessel coefficients to obtain a narrow-band signal, which is decomposed further into amplitude modulated (AM) and frequency modulated (FM) components. The discrete energy separation algorithm (DESA) is used to compute the amplitude envelope (AE) of the narrow-band AM-FM signal. Events such as the consonant and vowel beginnings in an unvoiced stop consonant vowel (SCV) and the GCIs are derived by processing the AE of the signal. The proposed approach for the detection of the VOT using the Bessel expansion is shown to perform better than the conventional Fourier representation. The performance of the proposed GCI detection method using the Bessel series expansion is compared against some of the existing methods for various noise environments and signal-to-noise ratios.     

Improvement of Transmission Control Protocol for High Bandwidth Applications

Transmission control protocol (TCP) is the widely and dominantly used protocol in today’s internet. A very recent implementation of congestion control algorithm is BBR by Google. Bottleneck bandwidth and round-trip time (BBR) is a congestion control algorithm which is created with the aim of increasing throughput and reducing delay. The congestion control protocols mentioned previously try to determine congestion limits by filling router queues. BBR drains the router queues at the bottleneck by sending exactly at the bottleneck link rate. This is done by the BBR through pacing rate which infers the delivery rate of the receiver and uses this as the estimated bottleneck bandwidth. But when the data rate is high, in the startup phase itself pipe becomes full and leads to some degradation in the Access Point of wireless environments by inducing losses specific to this environment. So the current pacing rate is not suitable for producing higher throughputs. Therefore, in the proposed system named R-BBR, this startup gain should be lower than the current startup gain which eventually would reduce pacing rate to reduce queue pressure in the sink node during the startup phase. The startup phase of BBR is modified to solve the problem of pipe full under high data rate. R-BBR has been evaluated over a wide range of wired as well as wireless networks by varying different factors like startup gain, congestion window, and pacing rate. It is inferred that R-BBR performs better than BBR with significant performance improvement.

     

Improved vowel region detection from a continuous speech using post processing of vowel onset points and vowel end-points

Multimedia Tools and Applications - Vowels are produced with an open configuration of the vocal tract, without any audible friction. The acoustic signal is relatively loud with varying strength of...     

Combining evidences from excitation source and vocal tract system features for Indian language identification using deep neural networks

In this paper, a combination of excitation source information and vocal tract system information is explored for the task of language identification (LID). The excitation source information is represented by features extracted from linear prediction (LP) residual signal called the residual cepstral coefficients (RCC). Vocal tract system information is represented by the mel frequency cepstral coefficients (MFCC). In order to incorporate additional temporal information, shifted delta cepstra (SDC) are computed. An LID system is built using SDC over both MFCC and RCC features individually and evaluated based on their equal error rate (EER). Experiments have been performed on a dataset consisting of 13 Indian languages with about 115 h for training and 30 h for testing using a deep neural network (DNN), DNN with attention (DNN-WA) and a state-of-the-art i-vector system. DNN-WA outperforms the baseline i-vector system. An EER of 9.93 and 6.25% are achieved using RCC and MFCC features respectively. By combining evidence from both features using a late fusion mechanism, an EER of 5.76% is obtained. This result indicates the complementary nature of the excitation source information to that of the widely used vocal tract system information for the task of LID.