Differentiable Weighted Finite-State Transducers

By Awni Y. Hannun, Vineel Pratap, Jacob Kahn, Wei-Ning Hsu. 2020

31 citations recorded by the source corpus.

We introduce a framework for automatic differentiation with weighted finite-state transducers (WFSTs) allowing them to be used dynamically at training time. Through the separation of graphs from operations on graphs, this framework enables the exploration of new structured loss functions which in turn eases the encoding of prior knowledge into learning algorithms. We show how the framework can combine pruning and back-off in transition models with various sequence-level loss functions. We also show how to learn over the latent decomposition of phrases into word pieces. Finally, to demonstrate that WFSTs can be used in the interior of a deep neural network, we propose a convolutional WFST layer which maps lower-level representations to higher-level representations and can be used as a drop-in replacement for a traditional convolution. We validate these algorithms with experiments in handwriting recognition and speech recognition.

Subfields: Natural Language Processing, Computational Linguistics & Semantics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Unsupervised Speech Recognition (2021, background): “See Hannun (2017) and Hannun et al. (2020) Figure 3 for an illustration.”