Rodrigo Bleyer (Helsinki):
Digital speech: an overview from the spoken language to the GIF problem
Wednesday, 16.12.2015 16:15 im Raum M5
Abstract: ``Digital Speech Processing" refers to the study of a speech signal. Namely, these signals are processed in a digital representation, as for example, synthesis, analysis, enhancement, compression and recognition may refers to this process.
In this talk we are interested on solving and explaining to the audience the core problem known as ``Glottal Inverse Filtering", or shortly, GIF. In only a few words, this problem can be modelled by convolving a pressure function (input signal) with an impulse response function (filter).
During this talk we shall comment on three selected approaches for solving the underlying problem, but we focus on our recent deterministic approach called ``AM-GIF", where AM stands for alternating minimisation. Finally, numerical reconstructions of the glottal flow volume are given in order to positioning the new approach among well known methods.
Angelegt am Saturday, 12.12.2015 13:55 von Martin Burger
Geändert am Tuesday, 15.12.2015 12:16 von Martin Burger
[Edit | Vorlage]