Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Utveckling av komplexitetsbedömningsmodell med XGBoost för att anpassa text-till-talsyntes inom Robot-Assisted Language Learning
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2019 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Development of complexity assessment model for adapting text-to-speech synthesis within Robot-Assisted Language Learning (English)
Abstract [sv]

Den sociala roboten Furhat har visat sig vara relevant för Robot-Assisted Language Learning. I dagsläget har Furhat en konstant talhastighet, men i tidigare studier har det framkommit att den skulle gynnas av att ha en mer dynamisk text-till-talsyntes. I denna kandidatexamensuppsats har en modell skapats, i syfte att klassificera meningar på svenska efter komplexitet. Modellen användes för att avgöra vilka meningar som bör ha en lägre talhastighet, i syfte att öka deras begripligheten för elever som lär sig svenska som andraspråk. För att utvärdera modellen jämfördes komplexitetsbedömningen med loggar från tidigare konversationer med Furhat, såväl som enkäter som fylldes i av studenter i svenska som andraspråk

Abstract [en]

The social robot Furhat has previously shown to be of interest for Robot-Assisted Language Learning. Currently, Furhat employs a constant speaking rate. However, conclusions from previous work have shown that from a comprehensional point of view, it could benefit from having a dynamic text-to-speech synthesis. In this bachelor thesis, a model was created in order to classify Swedish sentences according to their complexity. The model was later used to decide which utterances were to have a slower speaking pace, in order to increase comprehension among Swedish as second language learners. To evaluate the model, complexity classifications were compared with logs from previous conversations held with Furhat, as well as having questionnaires filled out by language learners. While the model showed some promising results with regards to classification of sentences, the evaluation was inconclusive with regards to its use as a way of dynamically assessing sentence complexity and adjusting Furhat’s speech rate accordingly.

Place, publisher, year, edition, pages
2019. , p. 13
Series
TRITA-EECS-EX ; 2019:581
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-263693OAI: oai:DiVA.org:kth-263693DiVA, id: diva2:1368852
Examiners
Available from: 2019-11-22 Created: 2019-11-08 Last updated: 2019-11-22Bibliographically approved

Open Access in DiVA

fulltext(1188 kB)4 downloads
File information
File name FULLTEXT01.pdfFile size 1188 kBChecksum SHA-512
f5b56ef06534cd16d2c714bd10d1d872de219da3962e9ceef47bebf26a8cbac115d9efe4c080f8598adaf64bba5a0df6d0f2b93db59c72cc1dd6f4ddb93f7a3a
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 4 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 4 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf