Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
AmbiSep: Ambisonic-to-Ambisonic Reverberant Speech Separation Using Transformer Networks
Ist Teil von
2022 International Workshop on Acoustic Signal Enhancement (IWAENC), 2022, p.1-5
Ort / Verlag
IEEE
Erscheinungsjahr
2022
Quelle
IEEE Xplore / Electronic Library Online (IEL)
Beschreibungen/Notizen
Consider a multichannel Ambisonic recording containing a mixture of several reverberant speech signals. Retrieving the reverberant Ambisonic signals corresponding to the individual speech sources blindly from the mixture is a challenging task as it requires estimating multiple signal channels for each source. In this work, we propose AmbiSep, a deep neural network-based plane-wave domain masking approach to solve this task. The masking network uses learned feature representations and transformers in a triple-path processing configuration. We train and evaluate the proposed network architecture on a spatialized WSJ0-2mix dataset and show that the method achieves a multichannel scale-invariant signal-to-distortion ratio improvement of 17.7 dB on the blind test set while preserving the spatial characteristics of the separated sounds.