Streaming Anchor Loss: Augmenting Supervision with Temporal Significance

arXiv:2310.05886v2 Announce Type: replace
Abstract: Streaming neural network models for fast frame-wise responses to various speech and sensory signals are widely adopted on resource-constrained platforms. Hence, increasing the learning capacity of such streaming models (i.e., by adding more parameters) to improve the predictive power may not be viable for real-world tasks. In this work, we propose a new loss, Streaming Anchor Loss (SAL), to better utilize the given learning capacity by encouraging the model to learn more from essential frames. …

abstract anchor arxiv capacity cs.ai cs.cv cs.lg loss network neural network parameters platforms power predictive responses sensory significance speech streaming supervision tasks temporal type wise world