Evaluating audiovisual source separation in the context of video conferencing

Inan, Berkay; Cernak, Milos; Grabner, Helmut; Tukuljac, Helena Peic; Pena, Rodrigo C. G.; Ricaud, Benjamin

doi:10.21437/Interspeech.2019-2671

Publication type:	Conference paper
Type of review:	Peer review (publication)
Title:	Evaluating audiovisual source separation in the context of video conferencing
Authors:	Inan, Berkay Cernak, Milos Grabner, Helmut Tukuljac, Helena Peic Pena, Rodrigo C. G. Ricaud, Benjamin
et. al:	No
DOI:	10.21437/Interspeech.2019-2671
Proceedings:	Proceedings Interspeech 2019
Page(s):	4579
Pages to:	4583
Conference details:	Interspeech 2019, Graz, Austria, 15-19 September 2019
Issue Date:	2019
Publisher / Ed. Institution:	International Speech Communication Association (ISCA)
Language:	English
Subjects:	Speech enhancement; Source separation; Multi-modal; Aaudiovisual
Subject (DDC):	621.3: Electrical, communications, control engineering
Abstract:	Source separation involving mono-channel audio is a challenging problem, in particular for speech separation where source contributions overlap both in time and frequency. This task is of high interest for applications such as video conferencing. Recent progress in machine learning has shown that the combination of visual cues, coming from the video, can increase the source separation performance. Starting from a recently designed deep neural network, we assess its ability and robustness to separate the visible speakers’ speech from other interfering speeches or signals. We test it for different configuration of video recordings where the speaker’s face may not be fully visible. We also asses the performance of the network with respect to different sets of visual features from the speakers’ faces.
URI:	https://digitalcollection.zhaw.ch/handle/11475/18478
Fulltext version:	Published version
License (according to publishing contract):	Licence according to publishing contract
Departement:	School of Engineering
Organisational Unit:	Institute of Data Analysis and Process Design (IDP)
Appears in collections:	Publikationen School of Engineering

Files in This Item:

There are no files associated with this item.

Show full item record

Inan, B., Cernak, M., Grabner, H., Tukuljac, H. P., Pena, R. C. G., & Ricaud, B. (2019). Evaluating audiovisual source separation in the context of video conferencing [Conference paper]. Proceedings Interspeech 2019, 4579–4583. https://doi.org/10.21437/Interspeech.2019-2671

Inan, B. et al. (2019) ‘Evaluating audiovisual source separation in the context of video conferencing’, in Proceedings Interspeech 2019. International Speech Communication Association (ISCA), pp. 4579–4583. Available at: https://doi.org/10.21437/Interspeech.2019-2671.

B. Inan, M. Cernak, H. Grabner, H. P. Tukuljac, R. C. G. Pena, and B. Ricaud, “Evaluating audiovisual source separation in the context of video conferencing,” in Proceedings Interspeech 2019, 2019, pp. 4579–4583. doi: 10.21437/Interspeech.2019-2671.

INAN, Berkay, Milos CERNAK, Helmut GRABNER, Helena Peic TUKULJAC, Rodrigo C. G. PENA und Benjamin RICAUD, 2019. Evaluating audiovisual source separation in the context of video conferencing. In: Proceedings Interspeech 2019. Conference paper. International Speech Communication Association (ISCA). 2019. S. 4579–4583

Inan, Berkay, Milos Cernak, Helmut Grabner, Helena Peic Tukuljac, Rodrigo C. G. Pena, and Benjamin Ricaud. 2019. “Evaluating Audiovisual Source Separation in the Context of Video Conferencing.” Conference paper. In Proceedings Interspeech 2019, 4579–83. International Speech Communication Association (ISCA). https://doi.org/10.21437/Interspeech.2019-2671.

Inan, Berkay, et al. “Evaluating Audiovisual Source Separation in the Context of Video Conferencing.” Proceedings Interspeech 2019, International Speech Communication Association (ISCA), 2019, pp. 4579–83, https://doi.org/10.21437/Interspeech.2019-2671.