Let's go public! taking a spoken dialog system to the real world

Raux, Antoine; Langner, Brian; Bohus, Dan; Black, Alan W.; Eskénazi, Maxine

doi:10.21437/interspeech.2005-399

Cited by 130 publications

(55 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The final message returned to the user is selected according to a trained scoring mechanism. In contrast, typical conversational agents use at least 3 modules to produce a response: a natural language understanding machine, a dialog manager (often made of many sub-modules), and a natural language generator (Raux et al, 2005;Callejas and López-Cózar, 2005).…”

Section: Discussionmentioning

confidence: 99%

“…In the goal-oriented setting, the systems are explicitly built to solve a particular task (McGlashan et al, 1992;Aust et al, 1995;Gorin et al, 1997). They typically operate in well-defined domains and use rule-based or modular architectures (Rudnicky et al, 1999;Raux et al, 2005). Therefore, while being accurate in their specific domain, goal-oriented systems often lack flexibility (Simpson and Eraser, 1993).…”

Section: Overview Of Dialog Systemsmentioning

confidence: 99%

See 1 more Smart Citation

The RLLChatbot: a solution to the ConvAI challenge

Gontier,

Sinha,

Henderson

et al. 2018

Preprint

View full text Add to dashboard Cite

Current conversational systems can follow simple commands and answer basic questions, but they have difficulty maintaining coherent and open-ended conversations about specific topics. Competitions like the Conversational Intelligence (ConvAI) challenge are being organized to push the research development towards that goal. This article presents in detail the RLLChatbot that participated in the 2017 ConvAI challenge. The goal of this research is to better understand how current deep learning and reinforcement learning tools can be used to build a robust yet flexible open domain conversational agent. We provide a thorough description of how a dialog system can be built and trained from mostly public-domain datasets using an ensemble model. The first contribution of this work is a detailed description and analysis of different text generation models in addition to novel message ranking and selection methods. Moreover, a new open-source conversational dataset is presented. Training on this data significantly improves the recall@k score of the ranking and selection mechanisms compared to our baseline model responsible for selecting the message returned at each interaction.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Overview Of Dialog Systemsmentioning

confidence: 99%

The RLLChatbot: a solution to the ConvAI challenge

Gontier,

Sinha,

Henderson

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…This definition encompasses most language-related interactive systems, including those based on audible speech interactions (e.g., CMU's Let's Go! system for bus information (Raux et al, 2005)).…”

Section: Background On Conversational Aimentioning

confidence: 99%

Conversational AI Systems for Social Good: Opportunities and Challenges

Peng¹,

Huang²,

Wu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Conversational artificial intelligence (ConvAI) systems have attracted much academic and commercial attention recently, making significant progress on both fronts. However, little existing work discusses how these systems can be developed and deployed for social good. In this paper, we briefly review the progress the community has made towards better ConvAI systems and reflect on how existing technologies can help advance social good initiatives from various angles that are unique for ConvAI, or not yet become common knowledge in the community. We further discuss about the challenges ahead for ConvAI systems to better help us achieve these goals and highlight the risks involved in their development and deployment in the real world.

show abstract

“…Everyday users now benefit from powerful QA technologies in a range of consumer-facing applications including health (Jacquemart and Zweigenbaum, 2003;Luo et al, 2015;Abacha and Demner-Fushman, 2016;Kilicoglu et al, 2018;, privacy (Sathyendra et al, 2017;Harkous et al, 2018;Ravichander et al, 2019), personal finance (Alloatti et al, 2019), search (Yang, 2015;Bajaj et al, 2016;He et al, 2018;Kwiatkowski et al, 2019) and dialog agents (Dahl et al, 1994;Raux et al, 2005). Voice assistants such as Amazon Alexa 2 or Google Home 3 have brought natural language technologies to several million homes globally (Osborne, 2016;Jeffs, 2018).…”

Section: Introductionmentioning

confidence: 99%

NoiseQA: Challenge Set Evaluation for User-Centric Question Answering

Ravichander¹,

Dalmia²,

Ryskina³

et al. 2021

Preprint

View full text Add to dashboard Cite

When Question-Answering (QA) systems are deployed in the real world, users query them through a variety of interfaces, such as speaking to voice assistants, typing questions into a search engine, or even translating questions to languages supported by the QA system. While there has been significant community attention devoted to identifying correct answers in passages assuming a perfectly formed question, we show that components in the pipeline that precede an answering engine can introduce varied and considerable sources of error, and performance can degrade substantially based on these upstream noise sources even for powerful pre-trained QA models. We conclude that there is substantial room for progress before QA systems can be effectively deployed, highlight the need for QA evaluation to expand to consider real-world use, and hope that our findings will spur greater community interest in the issues that arise when our systems actually need to be of utility to humans. 1

show abstract

Let's go public! taking a spoken dialog system to the real world

Cited by 130 publications

References 8 publications

The RLLChatbot: a solution to the ConvAI challenge

The RLLChatbot: a solution to the ConvAI challenge

Conversational AI Systems for Social Good: Opportunities and Challenges

NoiseQA: Challenge Set Evaluation for User-Centric Question Answering

Contact Info

Product

Resources

About