This paper introduces a novel speech enhancement method based on Empirical Mode Decomposition (EMD) and softthresholding algorithms. A modified soft thresholding strategy is adapted to the intrinsic mode functions (IMF) of the noisy speech. Due to the characteristics of EMD, each obtained IMF of the noisy signal will have different noise and speech energy distribution, thus will have a differen...

This work compares the accuracy of fundamental frequency and formant frequency estimation methods and maximum a posteriori (MAP) prediction from MFCC vectors with hand-corrected references. Five fundamental frequency estimation methods are compared to fundamental frequency prediction from MFCC vectors in both clean and noisy speech. Similarly, three formant frequency estimation and prediction m...

In this paper, we propose a novel algorithm to enhance the noisy speech in the framework of dual-channel speech enhancement. The new method is a hybrid optimization algorithm, which employs the  combination of  the  conventional θ-PSO and the shuffled sub-swarms particle optimization (SSPSO) technique. It is known that the θ-PSO algorithm has better optimization performance than standard PSO al...

In this paper, an improved mismatch function by considering signal correlation between speech and noise is proposed to better estimate the noisy speech HMM’s. A linearized model based on Taylor series expansion approach is used to approximate the proposed mismatch function. The parameters of the noisy speech HMM’s can be estimated more precisely by combining the parameters of the clean speech a...

In Speech processing Signal to Noise (SNR )Ratio plays an important role for Noise cancellation. here an attempt is made to evaluate SNR for given Noisy Speech signal using adaptive technique Binary LMS algorithm. In this proposal , Introduces the two control parameters of the Adaptive Algorithm rather than one control parameter as in case of LMS Algorithm. Also comparirison is made with respec...

Voice activity detection (VAD) is mainly used to detect speech/non-speech periods in observed noisy signals. The detected periods are used to reduce noise components or enhance speech components in noisy speech. However, current VAD techniques have serious problems in that the accuracy of detection of speech/non-speech periods drastically reduces if they are used for noisy speech and/or for mix...

Many short-time Fourier transform (STFT) based singlechannel speech enhancement algorithms are focused on estimating clean speech spectral amplitude from the noisy observed signal in order to suppress the additive noise. To this end, the state-of-the-art speech enhancement algorithms, employ noisy amplitude information and correspondingly a priori and a posteriori SNRs while they use no informa...

We present a method for learning a low-dimensional manifold for speech from clean speech samples in high-dimensional space. Using this manifold, we perform speech denoising by projecting noisy speech onto the manifold to remove nonspeech components. This method of denoising classifies our algorithm as a signal subspace denoising method, where highdimensional noisy data is projected onto the sig...

In the paper is described the BG-SRDat (BulGarian language Speaker Recognition DATa) a corpus in Bulgarian language collected over noisy analog telephone channels and intended for speaker recognition. The BG-SRDat comprises two different speech data, called Speech Data 1 (SD1) and Speech Data 2 (SD2), respectively. The SD1 is a reading text from a newspaper and its average length is about 40 se...