We introduce LAST, a LAttice-based Speech Transducer library in JAX. With an emphasis on flexibility, ease-of-use, and scalability, LAST implements differentiable weighted finite state automaton (WFSA) algorithms needed for training \& inference that scale to a large WFSA such as a recognition lattice over the entire utterance. Despite these WFSA algorithms being well-known in the literature, new challenges arise from performance characteristics of modern architectures, and from nuances in automatic differentiation. We describe a suite of generally applicable techniques employed in LAST to address these challenges, and demonstrate their effectiveness with benchmarks on TPUv3 and V100 GPU.
@article{arxiv.2304.13134,
title = {LAST: Scalable Lattice-Based Speech Modelling in JAX},
author = {Ke Wu and Ehsan Variani and Tom Bagby and Michael Riley},
journal= {arXiv preprint arXiv:2304.13134},
year = {2023}
}