Deep neural network based guided speech bandwidth extension

Schmidt K, Edler B (2019)


Publication Type: Conference contribution

Publication year: 2019

Publisher: Audio Engineering Society

Conference Proceedings Title: 147th Audio Engineering Society International Convention 2019

Event location: New York, NY US

Abstract

Up to today telephone speech is still limited to the range of 200 to 3400 Hz since the predominant codecs in public switched telephone networks are AMR-NB, G.711 and G.722 [1, 2, 3]. Blind bandwidth extension (blind BWE, BBWE) can improve the perceived quality as well as the intelligibility of coded speech without changing the transmission network or the speech codec. The BBWE used in this work is based on deep neural networks (DNNs) and has already shown good performance [4]. Although this BBWE enhances the speech without producing too many artifacts it sometimes fails to enhance prominent fricatives which can result in muffled speech. In order to better synthesize prominent fricatives the BBWE is extended by sending a single bit of side information - here referred to as guided BWE. This bit may be transmitted e.g. by watermarking so that no changes to the transmission network or the speech codec have to be done. Different DNN configurations (including convolutional (Conv.) layers as well as long short-term memory layers (LSTM)) making use of this bit have been evaluated. The BBWE has a low computational complexity and an algorithmic delay of 12 ms only and can be applied in state-of-the-art speech and audio codecs.

Authors with CRIS profile

How to cite

APA:

Schmidt, K., & Edler, B. (2019). Deep neural network based guided speech bandwidth extension. In 147th Audio Engineering Society International Convention 2019. New York, NY, US: Audio Engineering Society.

MLA:

Schmidt, Konstantin, and Bernd Edler. "Deep neural network based guided speech bandwidth extension." Proceedings of the 147th Audio Engineering Society International Convention 2019, New York, NY Audio Engineering Society, 2019.

BibTeX: Download