Joint Transformer/RNN Architecture for Gesture Typing in Indic Languages

Proceedings of the 28th International Conference on Computational Linguistics, pages 999–1010 Barcelona, Spain (Online), December 8-13, 2020 999 Joint Transformer/RNN Architecture for Gesture Typing in Indic Languages Emil Biju, Anirudh Sriram, Mitesh M. Khapra, Pratyush Kumar Indian Institute of Technology Madras {ee17b071,ee18b073}@smail.iitm.ac.in miteshk@cse.iitm.ac.in pratyushkpanda@gmail.com Abstract Gesture typing is a method of typing words on a touch-based keyboard by creating a continuous trace passing through the relevant keys. This work is aimed at developing a keyboard that supports gesture typing in Indic languages. We begin by noting that when dealing with Indic languages, one needs to cater to two different sets of users: (i) users who prefer to type in the native Indic script (Devanagari, Bengali, etc.) and (ii) users who prefer to type in the English script but want the transliterated output in the native script. In both cases, we need a model that takes a trace as input and maps it to the intended word. To enable the development of these models, we create and release two datasets. First, we create a dataset containing keyboard traces for 193,658 words from 7 Indic languages. Second, we curate 104,412 English-Indic transliteration pairs from Wikidata across these languages. Using these datasets we build a model that performs path decoding, transliteration and transliteration correction. Unlike prior approaches, our proposed model does not make co-character independence assumptions during decoding. The overall accuracy of our model across the 7 languages varies from 70-95%. 1 Introduction Gesture typing (Yang et al., 2019) is a fast and convenient method for providing textual input to a touch- based keyboard on small-sized devices like mobile phones. The conventional method of prodding the keys to input a word requires several discrete key presses resulting in impeded performance (Zhai and Kristensson, 2012). Besides, touch-based keyboards do not provide the rich tactile sensation of key boundaries and key presses that mitigates erroneous typing on a regular desktop keyboard (Kristensson, 2007). Gesture typing allows the user to input a word by drawing a single continuous trace over the keyboard and the finger needs to be lifted only once a word is completed. It takes advantage of muscle memory (Yang et al., 2019) which allows the user to subconsciously memorise input shapes for common words, resulting in improved typing speed and accuracy. The advantages of gesture typing over conventional typing methods are of particular significance in Indic language keyboards as Indic languages have a larger character set than English (for example,the Hindi script contains 52 alphabets as opposed to 26 in English), thus making the keyboard denser and increasing the chances of prodding the wrong character. A solution to this is to use an English character keyboard which receives an input sequence of English characters phonetically resembling the Indic word and transliterates them into the Indic language word (Hellsten et al., 2017). Besides, prior experience with using English keyboards ensures that minimal effort is elicited from the user while switching to the new keyboard. Therefore, in this work, we propose solutions to two tasks: …

Content truncated to 500 words