CoChat

The DOI matches its Crossref record; 74 of the 74 references it cites were found in a registry or an index; 28 of the 158 sources its claims cite were checked for support.

System and User Strategies to Repair Conversational Breakdowns of Spoken Dialogue Systems: A Scoping Review

Open access
Alghamdi, Essam, Halvey, Martin, Nicol, Emma · 2024 · ACM Conversational User Interfaces 2024

Abstract

Spoken Dialogue Systems (SDSs) are critical in facilitating natural and efficient human-machine interaction through speech. SDSs frequently encounter challenges in managing complex dialogues, resulting in communication breakdowns, which include misunderstandings— where the system misunderstands user input— and non-understandings— where the system fails to interpret the input at all. Strategies to repair these breakdowns have been investigated across multiple disciplines; despite this interest, the findings from these studies are inconsistent and hinder comparative analysis due to the use of diverse methodologies and terminologies. To address this gap, this scoping review systematically examines SDS and user repair strategies within a broad spectrum of literature. Based on 36 papers out of 818 found, we provide two comprehensive frameworks: one categorising SDS system-repair strategies into six distinct categories and the other user-repair strategies into five categories. Our analysis reveals a disparity in the literature’s focus on repair strategies, highlighting, in particular, the lack of research on less explored strategies, such as Information and Disclosure repair strategies, providing potential avenues for future research directions in this area.

Overview

CoChat Summary

TL;DR

This scoping review synthesizes 36 papers, selected from 818 found, on repair strategies for conversational breakdowns in spoken dialogue systems (SDSs). It proposes frameworks categorizing system-repair strategies into six categories and user-repair strategies into five categories, while highlighting uneven research coverage and terminology inconsistencies.

Key findings

  • From 36 included papers, system-repair strategies are categorized into six types: confirmation, information, social, solve, ask, and disclosure; user-repair strategies into five: clarification, information adjustment, modulation, system-centric, and user adaptation.
  • Most system-repair studies focused on confirmation, asking, and social strategies; only N=3 studies focused on information repair strategies, and there was a complete absence of studies investigating disclosure as a repair strategy.
  • For user-repair, clarification and information strategies were more common, while modulation, system-centric, and user adaptation strategies were notably under-explored; clarification strategies such as repetition and reformulation were the most widely employed, but repetition and modulation were often least successful and could negatively impact SDS recognition.
  • Methodologically, the majority of studies were mixed-method (N=16), followed by qualitative (N=10) and quantitative (N=10); field-collected corpora (N=11) and lab-collected corpora (N=10) were common, as were between-subject (N=10) and within-subject (N=10) experiments. Sample sizes ranged from 5 to 850 participants; English was the most prevalent language (N=16) and the United States the most common location (N=16).

Methods

  • Scoping review following PRISMA-ScR and Arksey & O’Malley’s framework; searched ACM Digital Library, IEEE Xplore, Scopus, and Web of Science up to August 2023, yielding 818 papers, reduced to 505 after duplicate removal, then 128 after title/abstract screening, 17 eligible from database search, plus 19 from backward/forward chaining, for 36 included papers.
  • Inclusion criteria: user studies examining strategies to repair or mitigate breakdowns in SDSs; voice-based interactions primarily; at least one user and one SDS; original peer-reviewed journal or full-length conference papers. Exclusion criteria included non-English publications, multimodal interactions where voice was not primary, purely technical error-handling studies, and dissertations/review articles/conference abstracts.
  • Data were charted and strategies categorized by function and objective, adapting existing frameworks where applicable; studies were classified as qualitative, quantitative, or mixed-method.

Limitations

  • The authors state that reliance on predefined frameworks may not fully account for the multifaceted nature of repair strategies; categorization was subjective and compounded by terminology inconsistencies; combined repair strategies blurred category distinctions; and the review was limited to four databases, predefined search terms, and inclusion criteria.
  • The authors also state that the review aimed to map strategies rather than evaluate their effectiveness on user experience or SDS performance.
  • Evident from the text: many field-collected corpora were pre-2009 and telephone-based; studies were predominantly Western and English-language; most examined single-session, single-language, disembodied SDS interactions, limiting generalizability to current and culturally diverse SDS contexts.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1145/3640794.3665558

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

22 Valid · 6 Not valid · 54 Not checked · 73 Own results and interpretations

65 more claims · 1 with their cited sources checked

Create a free account to see all 155 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 48
  • The quote only partly states this claim. 2
  • Probably not a substantial claim. 53
  • Borderline: one of the tests for a substantial claim is unclear. 2

Create a free account to see these claims.

Citations

Cites

Research works printed but not linked: 6

  1. Dennis Benner, Edona Elshan, Sofia Schöbel, and Andreas Janson. 2021. What do you mean? A review on recovery strategies to overcome conversational breakdowns of conversational agents. In International Conference on Information Systems (ICIS). 1–17.
  2. Stephen Choularton and Robert Dale. 2004. User responses to speech recognition errors: consistency of behaviour across domains. In Proceedings of SST, Vol. 4.
  3. Lea Kisser and Ingo Siegert. 2022. Erroneous reactions of voice assistants" in the wild"–first analyses. Elektronische Sprachsignalverarbeitung (2022), 113–120.
  4. 10.1080/10447318.2023.2266791 · DOI · Paper page
  5. Gabriel Skantze. 2003. Exploring human error handling strategies: Implications for spoken dialogue systems. In ISCA Tutorial and Research Workshop on Error Handling in Spoken Dialogue Systems.
  6. Lionel Sujay Vailshery. 2021. Number of digital voice assistants in use worldwide 2019-2024. Statista Research Departement.[Online]. Available: https://www. statista. com/statistics/973815/worldwide-digital-voice-assistant-in-use (2021).

Cited by

See all 36

Metadata

Authors
Alghamdi, E., Halvey, M., & Nicol, E.
Affiliations
University of Strathclyde (GB)
Venue
ACM Conversational User Interfaces 2024
Publisher
ACM
Published
2024-07-08
Pages
1-13
OpenAlex
W4400396839
Cited by
36
References
59
Citations by year
2026: 27 · 2025: 8 · 2024: 1
FWCI
10.61
Type
conference paper
Language
en
Open access
Locations
  • ACM Conference on Conversational User Interfaces (CUI) (publishedVersion, cc-by)
  • Strathprints: The University of Strathclyde institutional repository (University of Strathclyde) (submittedVersion)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
  • Social Sciences › Business, Management and Accounting › Marketing › AI in Service Interactions
  • Social Sciences › Arts and Humanities › Language and Linguistics › Language, Discourse, Communication Strategies
Keywords
spoken dialogue systems, conversational breakdown, repair strategies, misunderstanding, non-understanding, human-machine interaction, dialogue management

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.