An Embedded Deep Learning based Word Prediction

TL;DR

This paper introduces a compact deep learning model for on-device word prediction that optimizes memory usage and prediction speed, enabling real-time performance on mobile devices.

Contribution

It presents a novel embedded deep learning approach for word prediction that reduces model size and improves prediction efficiency compared to existing methods.

Findings

Model size is 7.40MB.

Average prediction time is 6.47 ms.

Improves keystroke savings and prediction rate.

Abstract

Recent developments in deep learning with application to language modeling have led to success in tasks of text processing, summarizing and machine translation. However, deploying huge language models for mobile device such as on-device keyboards poses computation as a bottle-neck due to their puny computation capacities. In this work we propose an embedded deep learning based word prediction method that optimizes run-time memory and also provides a real time prediction environment. Our model size is 7.40MB and has average prediction time of 6.47 ms. We improve over the existing methods for word prediction in terms of key stroke savings and word prediction rate.