Exploring Strategies for Modeling Sign Language Phonology

TL;DR

This paper investigates different learning strategies, including curriculum learning, for modeling sign language phonemes using graph convolution networks, achieving high accuracy on the Sem-Lex Benchmark.

Contribution

It introduces the application of curriculum learning to sign language phoneme recognition, demonstrating improved performance over other strategies.

Findings

Curriculum learning achieves 87% average accuracy on phoneme recognition.

Graph convolution networks effectively model sign language phonemes.

Curriculum learning outperforms fine-tuning and multi-task approaches.

Abstract

Like speech, signs are composed of discrete, recombinable features called phonemes. Prior work shows that models which can recognize phonemes are better at sign recognition, motivating deeper exploration into strategies for modeling sign language phonemes. In this work, we learn graph convolution networks to recognize the sixteen phoneme "types" found in ASL-LEX 2.0. Specifically, we explore how learning strategies like multi-task and curriculum learning can leverage mutually useful information between phoneme types to facilitate better modeling of sign language phonemes. Results on the Sem-Lex Benchmark show that curriculum learning yields an average accuracy of 87% across all phoneme types, outperforming fine-tuning and multi-task strategies for most phoneme types.