Abstract: Recent advances in neural text-to-speech (TTS) systems have achieved high-quality speech synthesis, but maintaining naturalness while controlling utterance duration remains a challenging ...

Abstract: Very short response time is a critical requirement for automatic discrete utterance recognition. The real-time vocabulary size of most of today's commercially available recognizers is ...

Dataset for ACL 2019 paper "Improving Multi-turn Dialogue Modelling with Utterance ReWriter " After another two months of human labeling, we release a much more ...

1 School of Translation Studies, Qufu Normal University, Rizhao, China 2 Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University, Kowloon, Hong Kong Although simultaneous ...

The positive dataset used in the ACL 2019 paper, 20000 dialogs. Each line in corpus.txt consists of four utterances of dialog (two context utterances, current utterance), and the rewritten uterance.

Increased communication rate has long been a goal of both individuals who use AAC and device manufacturers. There is evidence that utterance-based approaches have the potential to deliver faster rates ...

Function words, especially frequently occurring ones such as (the, that, and, and of), vary widely in pronunciation. Understanding this variation is essential both for cognitive modeling of lexical ...

1 Department of Psychology, California State University Channel Islands, Camarillo, CA, United States 2 Department of Psychology, University of California, Santa Cruz, Santa Cruz, CA, United States In ...

This is a hybrid seminar. To register for in-person attendance at SHHK Conference Room (SHHK Level 5), register here: https://wis.ntu.edu.sg/pls/webexe88/REGISTER_NTU ...