An extension of slow feature analysis for nonlinear blind source separation

We present and test an extension of slow feature analysis as a novel approach to nonlinear blind source separation. The algorithm relies on temporal correlations and iteratively reconstructs a set of statistically independent sources from arbitrary nonlinear instantaneous mixtures. Simulations show that it is able to invert a complicated nonlinear mixture of two audio signals with a high reliability. The algorithm is based on a mathematical analysis of slow feature analysis for the case of input data that are generated from statistically independent sources. © 2014 Henning Sprekeler, Tiziano Zito and Laurenz Wiskott.

Cell shape changes indicate a role for extrinsic tensile forces in Drosophila germ-band extension.

Drosophila germ-band extension (GBE) is an example of the convergence and extension movements that elongate and narrow embryonic tissues. To understand the collective cell behaviours underlying tissue morphogenesis, we have continuously quantified cell intercalation and cell shape change during GBE. We show that the fast, early phase of GBE depends on cell shape change in addition to cell intercalation. In antero-posterior patterning mutants such as those for the gap gene Krüppel, defective polarized cell intercalation is compensated for by an increase in antero-posterior cell elongation, such that the initial rate of extension remains the same. Spatio-temporal patterns of cell behaviours indicate that an antero-posterior tensile force deforms the germ band, causing the cells to change shape passively. The rate of antero-posterior cell elongation is reduced in twist mutant embryos, which lack mesoderm. We propose that cell shape change contributing to germ-band extension is a passive response to mechanical forces caused by the invaginating mesoderm.

Exploiting Chinese character models to improve speech recognition performance

The Chinese language is based on characters which are syllabic in nature. Since languages have syllabotactic rules which govern the construction of syllables and their allowed sequences, Chinese character sequence models can be used as a first level approximation of allowed syllable sequences. N-gram character sequence models were trained on 4.3 billion characters. Characters are used as a first level recognition unit with multiple pronunciations per character. For comparison the CU-HTK Mandarin word based system was used to recognize words which were then converted to character sequences. The character only system error rates for one best recognition were slightly worse than word based character recognition. However combining the two systems using log-linear combination gives better results than either system separately. An equally weighted combination gave consistent CER gains of 0.1-0.2% absolute over the word based standard system. Copyright © 2009 ISCA.