AI RESEARCH

ECHO: Frequency-aware Hierarchical Encoding for Variable-length Signals

arXiv CS.LG

ArXi:2508.14689v4 Announce Type: replace-cross Pre-trained foundation models have nstrated remarkable success in audio, vision and language, yet their potential for general machine signal modeling with arbitrary sampling rates-covering acoustic, vibration, and other industrial sensor data-remains under-explored. In this work, we propose a novel foundation model ECHO that integrates an advanced band-split architecture with frequency positional embeddings, enabling spectral localization across arbitrary sampling configurations.