People

Lab Director and Principle Investigator

Dr Andrew Hines

As well as leading QxLab, Andrew is the Director of Research for the UCD School of Computer Science. His research interests include machine learning, multimedia quality of experience, models for quality of life, cryptosystems and machine ethics. He has collaborated closely with Google since 2012 and currently has research projects funded by Science Foundation Ireland, European H2020, the Irish Research Council and Google. He has active collaborations with the Alan Turing Institute, Queen Mary University of London and Universidade de Brasília. He was general co-chair of the IEEE sponsored QoMEX 2020 Conference.

Postdocs

Dr Helard Becerra

Helard is a Postdoctoral Research Fellow at QxLab, University College Dublin. He received BS degree in computer science from Universidad Nacional San Antonio Abad del Cusco (UNSAAC), Peru, in 2010. He has completed MSc and PhD degrees in computer science from the University of Brasília, Brazil, in 2013 and 2019. Helard was a visiting researcher between 2017 and 2018 at University College Dublin (UCD, Dublin). He participated in research conducted at Samsung R&D Institute Brazil, in 2019.

Research interests: audio-visual signals, quality metrics, multimedia processing and machine learning.

PhD Researchers

Alessandro Ragano

Alessandro’s research is focused on the quality of experience for audio archives. He is using data driven approaches and machine learning to predict the QoE for archive speech and music archive material. His SFI sponsored research is in partnership with the Alan Turing Institute in London and he is cosupervised by Emmanouil Benetos from Queen Mary University of London.

Daniel Becker

Daniel Becker is currently working part-time on a Ph.D. in Cryptographic protocols. For the rest of his time, Daniel leads a global Engineering team Red Hat. Prior to joining Red Hat  5 years ago, Daniel held a variety of engineering leadership roles for companies in a venture capital/incubator group over the last 20 years. Daniel has a Computer Engineering degree from UL and an MBA from UCD.

Pheobe Sun

Pheobe is a PhD student in computer science doing research on speech and audio quality of experience, with an emphasis on cognitive-perceptual modelling. She gained her primary degree in social science and a master’s degree in psychoacoustics in Trinity College Dublin prior to joining QxLab.
 
 
 
Diptasree Debnath
 
Diptasree joined the lab (Sept 2022) as a PhD student working on Multi-dimensional Quality Assessment of Synthesized Speech. This research will be co-funded by SFI Insight and Xperi Corp. In the evaluation of Text-To-Speech (TTS) systems, one of the main performance indicators is how natural the synthesized speech sounds. To measure the naturalness, auditory listening tests are performed. In these tests, participants rate the naturalness on a five-point absolute category scale. The average across all test participants then gives the naturalness MOS (mean opinion score). Although this evaluation is time-consuming and costly and there is no reliable instrumental model available. The project will explore the potential of self supervised data driven models (e.g. wav2vec) to learn speech representations that can measure the quality and naturalness of synthetic speech.
 
Asad Ullah 
 
Asad joined QxLab in September 2022 as a PhD student working on Intermediate Speech Representations for Low Resource Speech Models. Asad will be part of the SFI ML-Labs cohort four. Feature extraction and intermediate speech representation are important components for speech processing tasks.  This project will adapt state-of-the-art deep learning architecture to improve the existing speech representation methods. New methods emerging from the fields of computer vision (CV) and natural language processing (NLP) will be explored for cross-domain inspiration.
 

Co-supervised Research Students

Davoud Shariat Panah

Davoud is a PhD student based in TU Dublin and is co-supervised with Dr Susan McKeever as part of the SFI D-Real Centre for Research Training.  His project is called SoundBeat: Heart Beat Sound Investigation using Deep Learning Neural Networks. He is using deep neural networks to train models for heart beat analysis and prediction using acoustic signals.

Adrielle Nazar Moraes

Adrielle is a research masters student based in Athlone Institute of Technology. She is exploring the quality of experience associated with spatial sound localisation in virtual reality applications.

Interns

Khushboo Satpute

Khushboo is pursuing her master’s in computer science with 6+ years of experience working as an Academician. She is currently working in the field of Data Analytics & Machine learning. Currently, She is working on Google Cloud Platform and Dialog flow.

 

Qijian Zhang

Qijian has completed his Master’s in Computer Science from UCD. He has 2 years of experience in front-end. Qijian is currently working on the development of web-based listening and speech test platform at QxLab.

 

Interns (2019)

Matthew Parker

Matthew developed a Unity application for VR experiments using an Oculus Quest to investigate the effect of the spatial audio on the perception of speech.

Interns (2018)

Feargus O’Gorman

Feargus is a back-end software developer with 5+ years of experience working on both enterprise SAAS projects and open-source projects. He has received his master degree UCD from School of Computer Science. Feargus spent a summer at QxLab collaborating with Google on software development for a speech quality model. He now works for Pivotal Software.

Tanvi Samel

Tanvi has completed her Master’s in Computer Science from UCD & Bachelor’s in Information Technology from Mumbai University. Tanvi carried out android application development at QxLab. She is currently working at SAP Ireland as SuccessFactors expert

 

Former QxLab

Dr Wissam A. Jassim

Wissam joined QxLab in 2019 based in the Insight Centre for Data Analytics. He worked in QxLab on robust speech quality models using machine learning in collaboration with Google developing the warp-q metric (ICASSP 2021 , IET 2022). He received his PhD degree in signals and systems from University of Malaya, Malaysia, in 2012. He joined the SFI Adapt Centre at Trinity College Dublin in 2016 and spent three years working on speech related research projects. Before his time at TCD, he was a postdoc fellow at University of Malaya from 2013 to 2016.

Research interests: Machine learning, neural responses of the human auditory periphery, speech processing applications, speech codecs, speech quality/intelligibility assessment, emotion speech, speech/speaker identification, and speech enhancement.

Dr Dan Barry

Dan is an audio signal processing engineer and was the CEO and co-founder of Riffstation which was acquired by Fender. He as developed audio processing algorithms for several companies in the past from startups to multinationals including Sony. He was a co-founder and manager of the Audio Research Group based in the Department of Electrical Engineering Systems at The Dublin Institute of Technology. His 2004 patented sound source separation technology was licensed to Sony and is still used in the PS3’s Singstar.

Areas of interest: Sound Source Separation, Spatial Audio, Audio Feature Extraction, VR

Dr Alcardo Alex Barakabitze

Alex completed his PhD in the University of Plymouth where he was part of the QoE-NET Marie Curie funded ITN PhD training network. His two year research project at QxLab was funded by a Government of Ireland Postdoctoral Fellowship (Irish Research Council) developing QoE centric management of multimedia services and orchestration of resources in 5G networks using SDN and NFV.

Dr Hamed Jahromi

Hamed has a background in networking and is a Cisco Certified Design Expert. He completed his PhD at QxLab conducting research on interactive web application QoE, He has received his B.Sc. in software engineering technologies. His current research interest includes network QoE management in Software Defined Networking (SDN), Internet traffic measurement, QoE/QoS, content-aware networking, Machine learning in communication networks. Hamed has over 15 years of industrial experience and worked as a principal network engineer. He has been involved in many large scale networking projects.

 

Dr AbuBakr Siddig

AbuBakr  was a post-doctoral researcher at UCD School of Computer Science, Ireland. He received his PhD degree in 2017 from University College Dublin, Ireland. He received his M.Sc. degree in 2010 from Blekinge Institute of Technology, Sweden. Dr Siddig obtained his B.Sc. (honours) degree in 2005 specialised in Electronics Engineering from the University of Khartoum, Sudan. He is a member of IEEE and a member of ACM. His current research focuses on the speech intelligibility in a VR environment.  Dr Siddig has vast experience in industry and academia. He was the business development manager at Canar Telecommunication, a fixed telecommunication services provider in Sudan, leading a team of professionals to deliver new products and services. He is currently a part-time lecturer at UCD school of computer science and a full-time lecturer at Griffith College, Ireland.  Dr Siddig research interest is signal processing and applied machine learning to Quality of Experience. He has an interest in other disciplines of computer science, particularly data science and chatbots.  His current research interests are QoE for Virtual Reality applications and personalised chatbots for eHealth.

Rahul Kumar Jaiswal, MSc (Research)

Rahul completed his research masters in quality of experience for realtime speech streaming in Summer 2019. He has now commenced a PhD at the WISENET lab in Norway.

 

Dr Colm Sloan

Colm worked on audio quality models in collaboration with YouTube. He now works at AI-based Vision analytics based company Artofus.

 

Dr Miroslaw (Mirek) Narbutt

Mirek worked on spatial audio compression analysis in collaboration with Google Chrome. He is now a lecturer in the Department of Electronic and Electrical Engineering at TU Dublin.