# LAST: Scalable Lattice-Based Speech Modelling in JAX

arXiv: https://arxiv.org/abs/2304.13134 · HTML: https://mlsanity.com/paper/2304.13134/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 4 of its 18 references that are in the map.
- First posted on arXiv 25 Apr 2023.

## Builds on (most cited first, 4 of 4)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Torch-Struct: Deep Structured Prediction Library](https://mlsanity.com/paper/2002.00876/) (2020)
- [Differentiable Weighted Finite-State Transducers](https://mlsanity.com/paper/2010.01003/) (2020)
- [Global Normalization for Streaming Speech Recognition in a Modular Framework](https://mlsanity.com/paper/2205.13674/) (2022)

## Abstract (arXiv)

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.