Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
From rule-based to data-driven lexical entrainment models in spoken dialog systems
INESC-ID Lisboa, Portugal; Instituto Superior Técnico, Portugal.ORCID-id: 0000-0002-8773-9216
CMU.
INESC-ID Lisboa/Instituto Superior Técnico.
2015 (engelsk)Inngår i: Computer speech & language (Print), ISSN 0885-2308, E-ISSN 1095-8363, Vol. 31, nr 1, s. 87-112Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

This paper presents uses a data-driven approach to improve Spoken Dialog System (SDS) performance by automatically finding the most appropriate terms to be used in system prompts. The literature shows that speakers use one another’s terms (entrain) when trying to create common ground during a spoken dialog. Those terms are commonly called “primes”, since they influence the interlocutors’ linguistic decision-making. This approach emulates human interaction, with a system built to propose primes to the user and accept the primes that the user proposes. These primes are chosen on the fly during the interaction, based on a set of features that indicate good candidate primes. A good candidate is one that we know is easily recognized by the speech recognizer, and is also a normal word choice given the context. The system is trained to follow the user’s choice of prime if system performance is not negatively affected. When system performance is affected, the system proposes a new prime. In our previous work we have shown how we can identify the prime candidates and how the system can select primes using rules. In this paper we go further, presenting a data-driven method to perform the same task. Live tests with this method show that use of on-the-fly entrainment reduces out-of-vocabulary and word error rate, and also increases the number of correctly transferred concepts.

sted, utgiver, år, opplag, sider
Academic Press, 2015. Vol. 31, nr 1, s. 87-112
Emneord [en]
Lexical entrainment; Spoken dialog systems; Data-driven model; Rule-based model
HSV kategori
Forskningsprogram
Människa-datorinteraktion; Informations- och kommunikationsteknik
Identifikatorer
URN: urn:nbn:se:kth:diva-186102DOI: 10.1016/j.csl.2014.11.007ISI: 000348959100005Scopus ID: 2-s2.0-84919924844OAI: oai:DiVA.org:kth-186102DiVA, id: diva2:925388
Merknad

QC 20160503

Tilgjengelig fra: 2016-05-02 Laget: 2016-05-02 Sist oppdatert: 2022-06-22bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Andre lenker

Forlagets fulltekstScopushttp://www.sciencedirect.com/science/article/pii/S0885230814001247

Person

Lopes, José

Søk i DiVA

Av forfatter/redaktør
Lopes, José
I samme tidsskrift
Computer speech & language (Print)

Søk utenfor DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric

doi
urn-nbn
Totalt: 103 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf