Language Models for Handwritten Short Message Services

TL;DR

This paper investigates handwriting recognition for SMS texts, focusing on unique linguistic phenomena like consonant skeletons, rebus, and phonetic writing, and evaluates language models' impact on recognition accuracy.

Contribution

It introduces a study of specific SMS language phenomena and compares recognition results with and without specialized language models.

Findings

Language models improve recognition accuracy for SMS phenomena

Consonant skeletons are challenging for standard recognition systems

Specialized models better handle rebus and phonetic writing

Abstract

Handwriting is an alternative method for entering texts composing Short Message Services. However, a whole new language features the texts which are produced. They include for instance abbreviations and other consonantal writing which sprung up for time saving and fashion. We have collected and processed a significant number of such handwriting SMS, and used various strategies to tackle this challenging area of handwriting recognition. We proposed to study more specifically three different phenomena: consonant skeleton, rebus, and phonetic writing. For each of them, we compare the rough results produced by a standard recognition system with those obtained when using a specific language model.