Development of a sound coding strategy based on a deep recurrent neural network for monaural source separation in cochlear implants

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Nogueira, Waldoca
• dc.contributor.author Gajecki, Tomca
• dc.contributor.author Krüger, Benjaminca
• dc.contributor.author Janer Mestres, Jordica
• dc.contributor.author Büchner, Andreasca
• dc.date.accessioned 2017-10-30T09:03:15Z
• dc.date.available 2017-10-30T09:03:15Z
• dc.date.issued 2016
• dc.description Comunicació presentada a la 12th ITG Conference on Speech Communication, celebrada els dies 5 a 7 d'octubre de 2016 a Paderborn, Alemanya.
• dc.description.abstract The aim of this study is to investigate whether a source separation algorithm based on a deep recurrent neural network (DRNN) can provide a speech perception benefit for cochlear implant users when speech signals are mixed with another competing voice. The DRNN is based on an existing architecture that is used in combination with an extra masking layer for optimization. The approach has been evaluated using the HSM sentence test (male voice) mixed with a competing voice (female voice) for a monaural speech separation task. Two DRNNs with two levels of complexity have been used. The algorithms have been evaluated in 8 normal hearing listeners using a Vocoder and in 3 CI users. Both DRNNs show a large and significant improvement in speech intelligibility using Vocoded speech. Preliminary results in 3 CI users seem to confirm the improvement observed using Vocoded simulations.en
• dc.description.sponsorship This work was supported by the DFG Cluster of Excellence EXC 1077/1 Hearing4all.
• dc.format.mimetype application/pdfca
• dc.identifier.citation Nogueira W, Gajecki T, Krueger B, Janer J, Buechner A. Development of a sound coding strategy based on a deep recurrent neural network for monaural source separation in cochlear implants. In: 12th ITG Conference on Speech Communication; 2016 Oct 5-7; Paderborn, Germany. Berlin (Germany): IEEE; 2016. p. 155-9.
• dc.identifier.uri http://hdl.handle.net/10230/33115
• dc.language.iso eng
• dc.publisher Institute of Electrical and Electronics Engineers (IEEE)ca
• dc.relation.ispartof 12th ITG Conference on Speech Communication; 2016 Oct 5-7; Paderborn, Germany. Berlin (Germany): IEEE; 2016. p. 155-9.
• dc.rights © 20165 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The final published article can be found at http://ieeexplore.ieee.org/document/7776166/
• dc.rights.accessRights info:eu-repo/semantics/openAccess
• dc.subject.other Música -- Anàlisi
• dc.title Development of a sound coding strategy based on a deep recurrent neural network for monaural source separation in cochlear implantsca
• dc.type info:eu-repo/semantics/conferenceObject
• dc.type.version info:eu-repo/semantics/acceptedVersion