Gmail smart compose: Real-time assisted writing

MX Chen, BN Lee, G Bansal, Y Cao, S Zhang… - Proceedings of the 25th …, 2019 - dl.acm.org
In this paper, we present Smart Compose, a novel system for generating interactive, real-
time suggestions in Gmail that assists users in writing mails by reducing repetitive ty**. In …

Evidence against the context-freeness of natural language

SM Shieber - The Formal complexity of natural language, 1985 - Springer
In searching for universal constraints on the class of natural languages, linguists have
investigated a number of formal properties, including that of context-freeness. Soon after …

Sub-lexical language models with word level pronunciation lexicons

H Sak, M Saraclar - US Patent 9,292,489, 2016 - Google Patents
BACKGROUND Automatic speech recognition (ASR) is an important tech nology that is
used in mobile devices to allow users to provide input as spoken words. In order to be …

Weighted automata algorithms

M Mohri - Handbook of weighted automata, 2009 - Springer
Weighted automata and transducers are widely used in modern applications in
bioinformatics and text, speech, and image processing. This chapter describes several …

Comparison of large language models in management advice for melanoma: Google's AI BARD, BingAI and ChatGPT

X Mu, B Lim, I Seth, Y **e, J Cevik… - Skin Health and …, 2024 - academic.oup.com
Abstract Background Large language models (LLMs) are emerging artificial intelligence (AI)
technology refining research and healthcare. Their use in medicine has seen numerous …

Phonetisaurus: Exploring grapheme-to-phoneme conversion with joint n-gram models in the WFST framework

JR Novak, N Minematsu, K Hirose - Natural Language Engineering, 2016 - cambridge.org
This paper provides an analysis of several practical issues related to the theory and
implementation of Grapheme-to-Phoneme (G2P) conversion systems utilizing the Weighted …

[КНИГА][B] Computational approaches to morphology and syntax

B Roark, R Sproat - 2007 - books.google.com
The book will appeal to scholars and advanced students of morphology, syntax,
computational linguistics and natural language processing (NLP). It provides a critical and …

Discriminative n-gram language modeling

B Roark, M Saraclar, M Collins - Computer Speech & Language, 2007 - Elsevier
This paper describes discriminative language modeling for a large vocabulary speech
recognition task. We contrast two parameter estimation methods: the perceptron algorithm …

[PDF][PDF] Discriminative language modeling with conditional random fields and the perceptron algorithm

B Roark, M Saraclar, M Collins… - Proceedings of the 42nd …, 2004 - aclanthology.org
This paper describes discriminative language modeling for a large vocabulary speech
recognition task. We contrast two parameter estimation methods: the perceptron algorithm …

Arabic diacritization using weighted finite-state transducers

R Nelken, S Shieber - Proceedings of the 2005 ACL Workshop on …, 2005 - dash.harvard.edu
Arabic is usually written without short vowels and additional diacritics, which are
nevertheless important for several applications. We present a novel algorithm for restoring …