Change search
ReferencesLink to record
Permanent link

Direct link
Using accent information in ASR models for Swedish
KTH, Superseded Departments, Speech, Music and Hearing.ORCID iD: 0000-0002-3323-5311
2003 (English)In: Proceedings of INTERSPEECH'2003, 2003, 2677-2680 p.Conference paper (Refereed)
Abstract [en]

In this study accent information is used in an attempt to improve acoustic models for automatic speech recognition (ASR). First, accent dependent Gaussian models were trained independently. The Bhattacharyya distance was then used in conjunction with agglomerative hierarchical clustering to define optimal strategies for merging those models. The resulting allophonic classes were analyzed and compared with the phonetic literature. Finally, accent "aware" models were built, in which the parametric complexity for each phoneme corresponds to the degree of variability across accent areas and to the amount of training data available for it. The models were compared to models with the same, but evenly spread, overall complexity showing in some cases a slight improvement in recognition accuracy.

Place, publisher, year, edition, pages
2003. 2677-2680 p.
National Category
Computer and Information Science
URN: urn:nbn:se:kth:diva-53571OAI: diva2:470361
QC 20111230Available from: 2011-12-28 Created: 2011-12-28 Last updated: 2011-12-30Bibliographically approved

Open Access in DiVA

No full text

Other links


Search in DiVA

By author/editor
Salvi, Giampiero
By organisation
Speech, Music and Hearing
Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
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

Total: 15 hits
ReferencesLink to record
Permanent link

Direct link