Advanced

Prediction of intonation patterns of accented words in a corpus of read Swedish news

Frid, Johan LU (2001) In Working Papers, Lund University, Dept. of Linguistics 49.
Abstract
This paper describes an initial attempt at the construction of a data-driven model of Swedish intonation. The study is mainly concerned with model building and prediction of the intonation patterns of accented words in a corpus of read news in Swedish. Extraction of pitch information is achieved by performing a stylization of the pitch contours. The information is used to build a model for the prediction of pitch patterns using linguistic features such as accent type and position of stress. The model is tested against unseen data from the same corpus. The evaluation is done by numerical comparisons. The RMSE between predicted and original contours for the different categories ranges between 3.7 and 31.4 Hz. The results are quite promising... (More)
This paper describes an initial attempt at the construction of a data-driven model of Swedish intonation. The study is mainly concerned with model building and prediction of the intonation patterns of accented words in a corpus of read news in Swedish. Extraction of pitch information is achieved by performing a stylization of the pitch contours. The information is used to build a model for the prediction of pitch patterns using linguistic features such as accent type and position of stress. The model is tested against unseen data from the same corpus. The evaluation is done by numerical comparisons. The RMSE between predicted and original contours for the different categories ranges between 3.7 and 31.4 Hz. The results are quite promising for future studies. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Working Paper
publication status
published
subject
in
Working Papers, Lund University, Dept. of Linguistics
volume
49
language
English
LU publication?
yes
id
b6012412-20d1-4f73-ad9d-aaa3057d25d3 (old id 528997)
alternative location
http://www.ling.lu.se/disseminations/pdf/49/bidrag11.pdf
date added to LUP
2007-09-27 13:15:59
date last changed
2017-10-02 06:30:01
@misc{b6012412-20d1-4f73-ad9d-aaa3057d25d3,
  abstract     = {This paper describes an initial attempt at the construction of a data-driven model of Swedish intonation. The study is mainly concerned with model building and prediction of the intonation patterns of accented words in a corpus of read news in Swedish. Extraction of pitch information is achieved by performing a stylization of the pitch contours. The information is used to build a model for the prediction of pitch patterns using linguistic features such as accent type and position of stress. The model is tested against unseen data from the same corpus. The evaluation is done by numerical comparisons. The RMSE between predicted and original contours for the different categories ranges between 3.7 and 31.4 Hz. The results are quite promising for future studies.},
  author       = {Frid, Johan},
  language     = {eng},
  note         = {Working Paper},
  series       = {Working Papers, Lund University, Dept. of Linguistics},
  title        = {Prediction of intonation patterns of accented words in a corpus of read Swedish news},
  volume       = {49},
  year         = {2001},
}