%Aigaion2 BibTeX export from Idiap Publications
%Sunday 22 December 2024 02:32:34 AM

@TECHREPORT{Cernak_Idiap-RR-07-2016,
         author = {Cernak, Milos and Asaei, Afsaneh and Bourlard, Herv{\'{e}}},
       keywords = {Binary pattern matching, Deep neural network (DNN) , Linguistic parsing, phonological posteriors , Structured sparse representation},
       projects = {Idiap},
          month = {4},
          title = {On Structured Sparsity of Phonological Posteriors for Linguistic Parsing},
           type = {Idiap-RR},
         number = {Idiap-RR-07-2016},
           year = {2016},
    institution = {Idiap},
            url = {http://arxiv.org/abs/1601.05647},
       abstract = {The speech signal conveys information on different time scales from short (20–40 ms) time scale or
segmental, associated to phonological and phonetic information to long (150–250 ms) time scale or supra
segmental, associated to syllabic and prosodic information. Linguistic and neurocognitive studies recognize
the phonological classes at segmental level as the essential and invariant representations used in speech
temporal organization.
In the context of speech processing, a deep neural network (DNN) is an effective computational method
to infer the probability of individual phonological classes from a short segment of speech signal. A vector of
all phonological class probabilities is referred to as phonological posterior. There are only very few classes
comprising a short term speech signal; hence, the phonological posterior is a sparse vector. Although
the phonological posteriors are estimated at segmental level, we claim that they convey supra-segmental
information. Namely, we demonstrate that phonological posteriors are indicative of syllabic and prosodic
events.
Building on findings from converging linguistic evidence on the gestural model of Articulatory Phonology
as well as neural basis of speech perception, we hypothesize that phonological posteriors convey properties
of linguistic classes at multiple time scales, and this information is embedded in their support (index) of
active coefficients. To verify this hypothesis, we obtain a binary representation of phonological posteriors at
segmental level which is referred to as first-order sparsity structure; the high-order structures are obtained
by concatenation of first-order binary vectors. It is then confirmed that classification of supra-segmental
linguistic events, the problem known as linguistic parsing, can be achieved with high accuracy using a simple
binary pattern matching of first-order or high-order structures.},
            pdf = {https://publications.idiap.ch/attachments/reports/2016/Cernak_Idiap-RR-07-2016.pdf}
}