Task Splitting for DNN-based Acoustic Echo and Noise Removal
Sebastian Braun, María Luis Valero
Abstract
Neural networks have led to tremendous performance gains for single-task speech enhancement, such as noise suppression and acoustic echo cancellation (AEC). In this work, we evaluate whether it is more useful to use a single joint or separate modules to tackle these problems. We describe different possible implementations and give insights into their performance and efficiency. We show that using a separate echo cancellation module and a module for noise and residual echo removal results in less near-end speech distortion and better performance during double-talk at same complexity.
Topics & Concepts
Echo (communications protocol)Computer scienceNoise (video)Task (project management)Speech recognitionDistortion (music)Speech enhancementImplementationResidualNoise reductionArtificial intelligenceAlgorithmTelecommunicationsEngineeringBandwidth (computing)Image (mathematics)Systems engineeringProgramming languageComputer networkAmplifierSpeech and Audio ProcessingAdvanced Adaptive Filtering TechniquesHearing Loss and Rehabilitation