{"date_created":"2019-07-12T05:27:39Z","citation":{"ama":"Drude L, Raj B, Haeb-Umbach R. On the appropriateness of complex-valued neural networks for speech enhancement. In: <i>INTERSPEECH 2016, San Francisco, USA</i>. ; 2016.","apa":"Drude, L., Raj, B., &#38; Haeb-Umbach, R. (2016). On the appropriateness of complex-valued neural networks for speech enhancement. In <i>INTERSPEECH 2016, San Francisco, USA</i>.","chicago":"Drude, Lukas, Bhiksha Raj, and Reinhold Haeb-Umbach. “On the Appropriateness of Complex-Valued Neural Networks for Speech Enhancement.” In <i>INTERSPEECH 2016, San Francisco, USA</i>, 2016.","ieee":"L. Drude, B. Raj, and R. Haeb-Umbach, “On the appropriateness of complex-valued neural networks for speech enhancement,” in <i>INTERSPEECH 2016, San Francisco, USA</i>, 2016.","mla":"Drude, Lukas, et al. “On the Appropriateness of Complex-Valued Neural Networks for Speech Enhancement.” <i>INTERSPEECH 2016, San Francisco, USA</i>, 2016.","short":"L. Drude, B. Raj, R. Haeb-Umbach, in: INTERSPEECH 2016, San Francisco, USA, 2016.","bibtex":"@inproceedings{Drude_Raj_Haeb-Umbach_2016, title={On the appropriateness of complex-valued neural networks for speech enhancement}, booktitle={INTERSPEECH 2016, San Francisco, USA}, author={Drude, Lukas and Raj, Bhiksha and Haeb-Umbach, Reinhold}, year={2016} }"},"status":"public","type":"conference","_id":"11756","oa":"1","publication":"INTERSPEECH 2016, San Francisco, USA","user_id":"44006","author":[{"last_name":"Drude","id":"11213","first_name":"Lukas","full_name":"Drude, Lukas"},{"first_name":"Bhiksha","full_name":"Raj, Bhiksha","last_name":"Raj"},{"last_name":"Haeb-Umbach","id":"242","first_name":"Reinhold","full_name":"Haeb-Umbach, Reinhold"}],"year":"2016","main_file_link":[{"open_access":"1","url":"https://groups.uni-paderborn.de/nt/pubs/2016/interspeech_2016_drude_paper.pdf"}],"language":[{"iso":"eng"}],"related_material":{"link":[{"url":"https://groups.uni-paderborn.de/nt/pubs/2016/interspeech_2016_drude_slides.pdf","relation":"supplementary_material","description":"Poster"}]},"title":"On the appropriateness of complex-valued neural networks for speech enhancement","department":[{"_id":"54"}],"abstract":[{"text":"Although complex-valued neural networks (CVNNs) â?? networks which can operate with complex arithmetic â?? have been around for a while, they have not been given reconsideration since the breakthrough of deep network architectures. This paper presents a critical assessment whether the novel tool set of deep neural networks (DNNs) should be extended to complex-valued arithmetic. Indeed, with DNNs making inroads in speech enhancement tasks, the use of complex-valued input data, specifically the short-time Fourier transform coefficients, is an obvious consideration. In particular when it comes to performing tasks that heavily rely on phase information, such as acoustic beamforming, complex-valued algorithms are omnipresent. In this contribution we recapitulate backpropagation in CVNNs, develop complex-valued network elements, such as the split-rectified non-linearity, and compare real- and complex-valued networks on a beamforming task. We find that CVNNs hardly provide a performance gain and conclude that the effort of developing the complex-valued counterparts of the building blocks of modern deep or recurrent neural networks can hardly be justified.","lang":"eng"}],"date_updated":"2022-01-06T06:51:08Z"}