16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

On Representation Learning for Artificial Bandwidth Extension

Matthias Zöhrer (1), Robert Peharz (2), Franz Pernkopf (1)

(1) Technische Universität Graz, Austria
(2) Medizinische Universität Graz, Austria

Recently, sum-product networks (SPNs) showed convincing results on the ill-posed task of artificial bandwidth extension (ABE). However, SPNs are just one type of many architectures which can be summarized as representational models. In this paper, using ABE as benchmark task, we perform a comparative study of Gauss Bernoulli restricted Boltzmann machines, conditional restricted Boltzmann machines, higher order contractive autoencoders, SPNs and generative stochastic networks (GSNs). Especially the latter ones are promising architectures in terms of its reconstruction capabilities. Our experiments show impressive results of GSNs, achieving on average an improvement of 3.90dB and 4.08dB in segmental SNR on a speaker dependent (SD) and speaker independent (SI) scenario compared to SPNs, respectively.

Full Paper

Bibliographic reference.  Zöhrer, Matthias / Peharz, Robert / Pernkopf, Franz (2015): "On representation learning for artificial bandwidth extension", In INTERSPEECH-2015, 791-795.