Zero-phase inverse filtering for extraction of source characteristics

The instants at which significant excitation of vocal tract take place during voicing are referred to as epochs. Epochs and strengths of excitation pulses at epochs are useful in characterizing voice source. Epoch filtering technique proposed by the authors determine epochs from speech waveform. In this paper we propose zero-phase inverse filtering to obtain strengths of excitation pulses at epochs. Zero-phase inverse filter compensates the gross spectral envelope of short-time spectrum of speech without affecting phase characteristics. Linear prediction analysis is used to realize the zero-phase inverse filter. Source characteristics that can be derived from speech using this technique are illustrated with examples.

A Scalable Architecture for VoIP Conferencing

Real-Time services are traditionally supported on circuit switched network. However, there is a need to port these services on packet switched network. Architecture for audio conferencing application over the Internet in the light of ITU-T H.323 recommendations is considered. In a conference, considering packets only from a set of selected clients can reduce speech quality degradation because mixing packets from all clients can lead to lack of speech clarity. A distributed algorithm and architecture for selecting clients for mixing is suggested here based on a new quantifier of the voice activity called “Loudness Number” (LN). The proposed system distributes the computation load and reduces the load on client terminals. The highlights of this architecture are scalability, bandwidth saving and speech quality enhancement. Client selection for playing out tries to mimic a physical conference where the most vocal participants attract more attention. The contributions of the paper are expected to aid H.323 recommendations implementations for Multipoint Processors (MP). A working prototype based on the proposed architecture is already functional.

A finite-rate-of-innovation signal sampling perspective of the source-filter model of speech production

We consider the speech production mechanism and the asso- ciated linear source-filter model. For voiced speech sounds in particular, the source/glottal excitation is modeled as a stream of impulses and the filter as a cascade of second-order resonators. We show that the process of sampling speech signals can be modeled as filtering a stream of Dirac impulses (a model for the excitation) with a kernel function (the vocal tract response),and then sampling uniformly. We show that the problem of esti- mating the excitation is equivalent to the problem of recovering a stream of Dirac impulses from samples of a filtered version. We present associated algorithms based on the annihilating filter and also make a comparison with the classical linear prediction technique, which is well known in speech analysis. Results on synthesized as well as natural speech data are presented.

CHARACTERIZATION AND HARDNESS OF Co-P COATINGS OBTAINED FROM DIRECT CURRENT ELECTRODEPOSITION USING GLUCONATE BATH

Direct current electrodeposition of Co-P alloy coatings were carried out using gluconate bath and they were characterized by employing techniques like XRD, FESEM, DSC and XPS. Broad XRD lines demonstrate the amorphous nature of Co-P coatings. Spherical and rough nodules are observed on the surface of coatings as seen from FESEM images. Three exothermic peaks around 290, 342 and 390 degrees C in DSC profiles of Co-P coatings could be attributed to the crystallization and formation of Co2P phase in the coatings. As-deposited coatings consist of Co metal and oxidized Co species as revealed by XPS studies. Bulk alloy P (P delta-) as well as oxidized P (P5+) are present on the surface of coatings. Concentrations of Co metal and P delta- increase with successive sputtering of the coating. Observed microhardness value is 1005 HK when Co-P coating obtained from 10 g L-1 NaH2PO2 is heated at 400 degrees C that is comparable with hard chromium coatings.

Autocatalytic duplex Ni-P/Ni-W-P coatings on AZ31B magnesium alloy

Autocatalytic duplex Ni-P/Ni-W-P coatings were deposited on AZ31B magnesium alloy using stabilizer free nickel carbonate bath. Some of the coated specimens were passivated in chromate solution with and without heat treatment. Plain Ni-P coatings were also prepared for comparison. Coatings were characterized for their surface morphology, composition and corrosion resistance. Energy dispersive analysis of X-ray (EDX) showed that the phosphorous content in the Ni-P coating is 6 wt.% and for Ni-W-P it reduced to 3 wt.% due to the codeposition of tungsten in the Ni-P coating. Marginal increase in P and W contents was observed on passivated coupons along with Cr (0.18 wt.%) and O (2.8 wt.%) contents. Field emission scanning electron microscopy (FESEM) examination of these coating surfaces exhibited the nodular morphology. Chromate passivated surfaces showed the presence of uniformly distributed bright Ni particles along with nodules. Potenfiodynamic polarization and electrochemical impedance spectroscopy (EIS) studies were carried out in deaerated 0.15 M NaCI solution to find out the corrosion resistance of the coatings. Among the coatings developed, duplex-heat treated-passivated (duplex-HIP) coatings showed lower corrosion current density (i(corr)) and higher polarization resistance (R-p) indicating the improved corrosion resistance. The charge transfer resistance (R-ct) value obtained for the duplex-HIP was about 170 times higher compared to that for Ni P coating. (c) 2013 Elsevier B.V. All rights reserved.

Do stable isotopes in carbonate cement of Mio-Pleistocene Himalayan sediments record paleoecological and paleoclimatic changes?

Non-pedogenic carbonates, such as carbonate cement and nodules in the sandstones, are quite common in the terrestrial geological record. Unlike pedogenic carbonates, their stable isotope ratios lack investigations for paleo-climatic reconstructions. The present investigation therefore, explores the possibility of use of stable isotope studies of non-pedogenic carbonates from the Mb-Pleistocene Siwalik Group of sediments exposed in the Ramnagar sub-basin of the NW Himalaya. Petrographic studies reveal the dominance of micrite fabric in carbonate nodules both of pedogenic and non-pedogenic samples irrespective of specific stratigraphic unit However, calcite as cement in the sandstones shows the dominance of micrite fabric in the younger in age sediments. Seventy-two non-pedogenic carbonate samples from the carbonate nodules and cement in the Siwalik sandstones, ranging in age between similar to 1 Ma and 12.2 Ma, were analyzed for delta C-13 and delta O-18 values. The delta C-13 values vary from -24.77 parts per thousand to -1.1 parts per thousand and delta O-18 values vary from -15.34 parts per thousand to -7.81 parts per thousand. Pedogenic and non-pedogenic carbonates ranging in age between similar to 1 Ma and 6 Ma have largely similar delta C-13 values and the range of delta C-13 values indicate the dominance of C-4 type of vegetation. However, unlike pedogenic carbonates which showed the dominance of C-3 type of vegetation pre- 7 Ma on the basis of delta C-13 -depleted isotopic values (Singh et al., 2011), delta C-13 values are largely enriched in the corresponding aged non-pedogenic carbonates revealing no information on specific type of vegetation. Likewise, paleoprecipitational reconstructions from delta O-18 values in pedogenic carbonates showed a progressive increase in aridity from similar to 12 Ma to recent excluding short term increases in rainfall/monsoon intensity at around 10 Ma, 5 Ma, and 1.8 Ma (Singh et al., 2012). On the contrary, such reconstructions are not possible from the delta O-18 values of non-pedogenic carbonates and indeed the delta O-18 values of non-pedogenic carbonates are largely depleted to as much as 6 parts per thousand from the corresponding pedogenic carbonates. Such differences in delta C-13 and delta O-18 values of non-pedogenic carbonates from pedogenic carbonates are primarily due to the dependence of the former on groundwater conditions responsible for precipitating carbonate. Further, a comparison of isotopic values between non-pedogenic and pedogenic carbonates can be interpreted that post-6 Ma and pre-6 Ma non-pedogenic carbonates were largely formed by shallow and deep groundwater conditions respectively. The result of these investigative studies therefore, suggests that the stable delta C-13 and delta O-18 values of non-pedogenic carbonates, unlike the pedogenic carbonates and irrespective of nature of calcite fabric, showed their little importance in paleoclimatic and paleoecological reconstructions. (C) 2014 Elsevier B.V. All rights reserved.

Demodulation of Narrowband Speech Spectrograms Using the Riesz Transform

We propose a two-dimensional (2-D) multicomponent amplitude-modulation, frequency-modulation (AM-FM) model for a spectrogram patch corresponding to voiced speech, and develop a new demodulation algorithm to effectively separate the AM, which is related to the vocal tract response, and the carrier, which is related to the excitation. The demodulation algorithm is based on the Riesz transform and is developed along the lines of Hilbert-transform-based demodulation for 1-D AM-FM signals. We compare the performance of the Riesz transform technique with that of the sinusoidal demodulation technique on real speech data. Experimental results show that the Riesz-transform-based demodulation technique represents spectrogram patches accurately. The spectrograms reconstructed from the demodulated AM and carrier are inverted and the corresponding speech signal is synthesized. The signal-to-noise ratio (SNR) of the reconstructed speech signal, with respect to clean speech, was found to be 2 to 4 dB higher in case of the Riesz transform technique than the sinusoidal demodulation technique.

An Alternating l(p) - l(2) Projections Algorithm (ALPA) for Speech Modeling using Sparsity Constraints

We address the problem of separating a speech signal into its excitation and vocal-tract filter components, which falls within the framework of blind deconvolution. Typically, the excitation in case of voiced speech is assumed to be sparse and the vocal-tract filter stable. We develop an alternating l(p) - l(2) projections algorithm (ALPA) to perform deconvolution taking into account these constraints. The algorithm is iterative, and alternates between two solution spaces. The initialization is based on the standard linear prediction decomposition of a speech signal into an autoregressive filter and prediction residue. In every iteration, a sparse excitation is estimated by optimizing an l(p)-norm-based cost and the vocal-tract filter is derived as a solution to a standard least-squares minimization problem. We validate the algorithm on voiced segments of natural speech signals and show applications to epoch estimation. We also present comparisons with state-of-the-art techniques and show that ALPA gives a sparser impulse-like excitation, where the impulses directly denote the epochs or instants of significant excitation.