José David Lopes

José David Lopes

Research Fellow

Heriot-Watt University

Biography

José David Lopes is a Research Fellow at Heriot Watt University in Edinburgh, Scotland. His research focus on conversational systems for human-robot interaction. He’s part of the Interaction Lab and the INCRA Lab. He is currently involved in The RobotCafé for the National Robotarium and the UKRI TAS Trust Node.

Download my resumé.

Interests

  • Conversational AI
  • Human-Robot Interaction
  • Natural Language Processing

Education

  • PhD in Electrical and Computer Engineering, 2013

    Instituto Superior Técnico

  • MEng in Electrical and Computer Engineering, 2008

    Universidade de Coimbra

  • BEng in Electrical and Computer Engineering, 2007

    Universidade de Coimbra

Experience

 
 
 
 
 

Research Fellow

Heriot-Watt University

Aug 2020 – Present Edinburgh, Scotland
Developing conversational interfaces for the oil and gas industry. RobotCafé, developing a robot-driven coffee machine using Rasa for the conversational part. Investigating signals of trust hidden in human-human and human-machine natural interactions.
 
 
 
 
 

Research Associate

Heriot-Watt University

Apr 2018 – Jul 2020 Edinburgh, Scotland
Conversational Interface Leader for the ORCA Hub: Developing an HCI system for operators to control robots on an off-shore facility. Used Tensorflow to train a neural network which models the conversation.
 
 
 
 
 

Post-doctoral Researcher

KTH Royal Institute of Technology

Mar 2014 – Mar 2018 Stockholm, Sweden
Worked on automatic error detectors for Interactive Voice Response systems, data-driven models for turn-taking in dialogues and a Social Robot to assist in a language café scenario.
 
 
 
 
 

Post-doctoral Researcher

KTH Royal Institute of Technology

Mar 2014 – Mar 2018 Stockholm, Sweden
Worked on automatic error detectors for Interactive Voice Response systems, data-driven models for turn-taking in dialogues and a Social Robot to assist in a language café scenario.
 
 
 
 
 

Research Engineer

University of the Basque Country

Oct 2013 – Feb 2014 Bilbao, Spain
Working on a conversational system for public transit schedule information for the city of Bilbao in Spanish. System was based on CMU’s Ravenclaw.
 
 
 
 
 

PhD Student/Researcher

INESC-ID Lisboa

Oct 2008 – Sep 2013 Lisbon, Portugal
Working on a conversational system for bus schedule information in Portuguese. System was based on CMU’s Ravenclaw. The system was used to study how phenomena in human communication transfer to conversational systems. Worked on Computer Aided Language Learning.
 
 
 
 
 

Junior Researcher

Instituto de Telecomunicações

Oct 2007 – Sep 2008 Coimbra, Portugal
Working on robust speech recognition for industrial environments as part of Tecnovoz project.

Featured Publications

Sarah Gillet, Ronald Cumbal, André Pereira, José David Lopes, Olov Engwall, Iolanda Leite
March 2021 In HRI21

Robot Gaze Can Mediate Participation Imbalance in Groups with Different Skill Levels (Best User Studies Paper)

Using a robot-head gaze to balance the a game interaction between a native and a non-native speaker.
PDF DOI
Francisco Javier Chiyah Garcia, Simon Smith, José David Lopes, Subramanian Ramamoorthy, Helen Hastie
March 2021 In HRI2021

Self-Explainable Robots in Remote Environments

Explaining remote robots behaviour learned from demonstration with natural language.
PDF DOI
Birthe Nesset, David A. Robnb, José David Lopes, Helen Hastie
March 2021 In HRI2021

Transparency in HRI: Trust and Decision Making in the Face of Robot Errors

A study on how robot-errors affect trust in the scenario of a COVID-19 triage.
PDF DOI
Rui Ribeiro, Alberto Abad, José David Lopes
February 2021 In IberSPEECH

Domain Adaptation in Dialogue Systems using Transfer and Meta-Learning

This article proposes meta-learning to perform domain adaptation in dialogue systems.
PDF
José David Lopes, Francisco Javier Chiyah Garcia, Helen Hastie
December 2020 In NeurIPS Workshop HLDS

The Lab vs The Crowd: An Investigation into Data Quality for Neural Dialogue Models

We compared the performance of neural models when trained using crowd-sourced and lab collected data. For our particular domain lab data lead to better performance.
Code Slides Video
Ronald Cumbal, José David Lopes, Olov Engwall
October 2020 In ICMI

Interaction and collaboration in robot-assisted language learning for adults

Analysing a developing uncertainty detectors for Robot-Aided Language Learning (RALL).
PDF DOI
Olov Engwall, José David Lopes
August 2020 In CALL

Interaction and collaboration in robot-assisted language learning for adults

This article analyses how robot–learner interaction in robot assisted language learning (RALL) is influenced by the interaction behaviour of the robot. Adaptation is needed to user preference and needs.
PDF
Francisco Javier Chiyah Garcia, José David Lopes, Xingkun Liu, Helen Hastie
May 2020 In LREC

CRWIZ: A Framework for Crowdsourcing Real-Time Wizard-of-Oz Dialogues

A framework to collect dialogues through crowd-sourcing using a Wizard-of-Oz setup..
PDF Code
Francisco Javier Chiyah Garcia, José David Lopes, Helen Hastie
May 2020 In MM-HRI

Natural Language Interaction to Facilitate Mental Models of Remote Robots

A position paper on mental models for HRI.
PDF
José David Lopes, David A. Robb, Xingkun Liu, Helen Hastie
March 2020 In HRI

Demonstration of a Social Robot for Control of Remote Autonomous Systems

Description of our social robot demo for the oil and gas industry.
PDF
Ronald Cumbal, José David Lopes, Olov Engwall
March 2020 In HRI2021

Uncertainty in Robot Assisted Second Language Conversation Practice

Preliminary results on uncertainty detection for Robot-Aided Language Learning (RALL).
PDF DOI
Olov Engwall, José David Lopes, Anna Åhlund
February 2020 In IJSR

Robot interaction styles for conversation practice in second language learning

This article analyses robot–learner interactions in robot assisted language learning (RALL) and proposes a set of recommendations to future studies.
PDF

Recent Publications

$Quickly discover relevant content by filtering publications.

Contact