Aug 2026
Multirate State Space Models for End-to-End Processing of Pulse Density Modulated Speech Signals
This paper proposes a novel end-to-end PDM speech processing architecture that uses an SSM to encode the input audio signal into a modulation- and sampling-rate-invariant latent representation, and achieves robust speech classification and enhancement gains at low-power sampling-rates.
Ludovic Boulanger, Sean U N Wood
· arXiv.org · 0 citations