View on GitHub

Aciel Eshky's homepage

Homepage

Introduction

I currently work as a Machine Learning Researcher at Rasa, where we are building the standard infrastructure for conversational AI.

Before joining Rasa, I was a Research Associate at the Centre for Speech Technology Research (CSTR). I worked on the Ultrax2020 project automating speech and language therapy using machine learning. I also supervised a number of Informatics MSc students.

Prior to that, I was a Senior Statistics Software Engineer at Quantics Biostatistics where I developed as part of a team a commercial software for the statistical modelling and analysis of Bioassay data (QuBAS). During my time there I also worked on a number of statistical consulting projects for pharmaceutical clients.

Before joining Quantics, I worked as a Researched Associate at the Institute of Perception, Action and Behaviour.

In 2014, I received my PhD from the Institute for Language, Cognition and Computation (ILCC) at the School of Informatics at the University of Edinburgh. My thesis focussed on modelling goal-directed users in task-oriented dialogs as generative probabilistic models. I was supervised by Mark Steedman, and co-supervised by Subramanian Ramamoorthy.

Publications

Full list on Google Scholar.

PhD Thesis

Generative Probabilistic Models of Goal-Directed Users in Task Oriented Dialogs. Aciel Eshky. PhD thesis, School of Informatics, University of Edinburgh, July 2014. [pdf] [bib]

Journal Articles

Automatic audiovisual synchronisation for ultrasound tongue imaging. Aciel Eshky, Joanne Cleland, Manuel Sam Ribeiro, Eleanor Sugden, Korin Richmond, Steve Renals (2021). In Speech Communication, Volume 132, Pages 83-95, ISSN 0167-6393. [DOI] [preprint]

Exploiting ultrasound tongue imaging for the automatic detection of speech articulation errors. Manuel Sam Ribeiro, Joanne Cleland, Aciel Eshky, Korin Richmond, Steve Renals (2021). In Speech Communication, Volume 128, Pages 24-34, ISSN 0167-6393. [DOI] [preprint]

Conference Articles

Silent versus modal multi-speaker speech recognition from ultrasound and video. Manuel Sam Ribeiro, Aciel Eshky, Korin Richmond, Steve Renals. INTERSPEECH 2021. [pdf]

TaL: a synchronised multi-speaker corpus of ultrasound tongue imaging, audio, and lip videos. Manuel Sam Ribeiro, Jennifer Sanger, Jing-Xuan Zhang, Aciel Eshky, Alan Wrench, Korin Richmond, Steve Renals. SLT 2021: IEEE Spoken Language Technology Workshop.[pdf]

Synchronising audio and ultrasound by learning cross-modal embeddings. Aciel Eshky, Manuel Sam Ribeiro, Korin Richmond, Steve Renals. In INTERSPEECH 2019: Proceedings of the 20th Annual Conference of the International Speech Communication Association (ISCA), Graz, Austria, 2019. [pdf] [poster]

Ultrasound tongue imaging for diarization and alignment of child speech therapy sessions. Manuel Sam Ribeiro, Aciel Eshky, Korin Richmond, Steve Renals. In INTERSPEECH 2019: Proceedings of the 20th Annual Conference of the International Speech Communication Association (ISCA), Graz, Austria, 2019. [pdf]

Speaker-Independent Classification of Phonetic Segments from Raw Ultrasound in Child Speech. Manuel Sam Ribeiro, Aciel Eshky, Korin Richmond, Steve Renals. In ICASSP 2019: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, United Kingdom, 2019. [pdf]

UltraSuite: A Repository of Ultrasound and Acoustic Data from Child Speech Therapy Sessions. Aciel Eshky, Manuel Sam Ribeiro, Joanne Cleland, Korin Richmond, Zoe Roxburgh, James Scobbie, Alan Wrench. In INTERSPEECH 2018: Proceedings of the 19th Annual Conference of the International Speech Communication Association (ISCA), Hyderabad, India, 2018. [pdf] [bib] [slides]

A Generative Model for User Simulation in a Spatial Navigation Domain. Aciel Eshky, Ben Allison, Subramanian Ramamoorthy, and Mark Steedman. In EACL: Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics (EACL), Gothenburg, pages 626-635. Association for Computational Linguistics, 2014. [pdf] [bib]

Generative Goal-Driven User Simulation for Dialog Management. Aciel Eshky, Ben Allison, and Mark Steedman. In EMNLP-CoNLL: Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, Jeju Island, Korea, 71-8. Association for Computational Linguistics, 2012. [pdf] [bib]

Code

UltraSuite tools: https://github.com/UltraSuite/ultrasuite-tools