A natural representation of data is given by the parameters which generated the data. If the space of parameters is continuous, then we can regard it as a manifold. In practice, w...
In this paper we describe a noise reduction preprocessing algorithm for the adaptive multirate (AMR) speech codec of the GSM system. The algorithm is based on spectral weighting a...
Peter Jax, Rainer Martin, Peter Vary, Marc Adrat, ...
Corruption of data by class-label noise is an important practical concern impacting many classification problems. Studies of data cleaning techniques often assume a uniform label ...
Conventional non-blind image deblurring algorithms involve natural image priors and maximum a-posteriori (MAP) estimation. As a consequence of MAP estimation, separate pre-process...
Previously we have proposed different models for estimating articulatory gestures and vocal tract variable (TV) trajectories from synthetic speech. We have shown that when deploye...
Vikramjit Mitra, Hosung Nam, Carol Y. Espy-Wilson,...