An effective nonlinear fourth-order partial differential equation (PDE) - based additive denoising technique is considered in this article. The proposed image restoration approach is based on a hybrid anisotropic diffusion model with boundary conditions, which combines nonlinear second and fourth order diffusion-based components and a conventional filter kernel. The obtained PDE model is well-posed under some certain assumptions and its weak unique solution is computed by applying a consistent and fast-converging finite difference method-based numerical approximation scheme that is also proposed here. The image denoising results produced by this method are finally described.
Voice activity detection (VAD), namely determining whether a speech signal is active or inactive, and single talk detector (STD), namely detecting that only one speaker is active, are important building blocks in many speech processing applications. A speaker-localization stage (such as the steered response power (SRP)) is often concurrently implemented on the same device.In this paper, the spatial properties of the SRP are utilized for improving the performance of both the voice activity detector (VAD) and the STD. We propose to measure the entropy at the SRP output and compare with the typical entropy of noise-only frames. This feature utilizes spatial information and may therefore become advantageous in nonstationary noise environments. The STD can then be implemented by determining local minimum values of the entropy measure of the SRP.The proposed VAD was tested for a single speaker with two cases, directional background noise with changing level and with a background music source. The proposed STD was tested using real recordings of two concurrent speakers.
The difference-to-sum power ratio was proposed and used to suppress wind noise under specific acoustic conditions. In this contribution, a general formulation of the difference-to-sum power ratio associated with a mixture of speech and wind noise is proposed and analyzed. In particular, it is assumed that the complex coherence of convective turbulence can be modelled by the Corcos model. In contrast to the work in which the power ratio was first presented, the employed Corcos model holds for every possible air stream direction and takes into account the lateral coherence decay rate. The obtained expression is subsequently validated with real data for a dual microphone set-up. Finally, the difference-to-sum power ratio is exploited as a spatial feature to indicate the frame-wise presence of wind noise, obtaining improved detection performance when compared to an existing multi-channel wind noise detection approach.
In this study, we aim to classify comments as abusive or non-abusive. We develop a Hebrew corpus of user comments annotated for abusive language. Then, we investigate highly sparse n-grams representations as well as denser character n-grams representations for comment abuse classification. Since the comments in social media are usually short, we also investigate four dimension reduction methods, which produce word vectors that collapse similar words into groups. We show that the character n-grams representations outperform all the other representation for the task of identifying abusive comments.
This paper considers a problem of lossy compression of generalized Gaussian (GG) sources (i.e., sources with the probability density functions proportional to $\mathrm {e}^{-\frac {|x|^{\mathrm {S}}}{2}}$, $s \gt 0$) with an $\ell _{r}, r \gt 0$, distortion measure.It is shown that an optimal reconstruction distribution always exists and properties of this distribution are studied. In particular, it is shown that if $s \leq r-1$ then an optimal reconstruction must have unbounded support and for $s \gt r$ an optimal reconstruction must have bounded support. Further, it is shown that Shannon’s lower bound is achievable if and only if $r = s \in (0,1$] $\cup \{2\}$, or in other words when the GG distribution is self-decomposable. Finally, conditions are shown under which an optimal reconstruction is discrete with finitely many mass points.
