Abstract
Segmenting the speech signals on the basis of time-frequency analysis is the most natural approach. Boundaries are located in places where energy of some frequency subband rapidly changes. Speech segmentation method which bases on discrete wavelet transform, the resulting power spectrum and its derivatives is presented. This information allows to locate the boundaries of phonemes. A statistical classification method was used to check which features are useful. The efficiency of segmentation was verified on a male speaker taken from a corpus of Polish language.
| Original language | English |
|---|---|
| Title of host publication | 2008 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOLS 1-4 |
| Place of Publication | NEW YORK |
| Publisher | IEEE |
| Pages | 1297-1300 |
| Number of pages | 4 |
| ISBN (Print) | 978-1-4244-2570-9 |
| Publication status | Published - 2008 |
| Event | IEEE International Conference on Multimedia and Expo (ICME 2008) - Hannover Duration: 23 Jun 2008 → 26 Jun 2008 |
Conference
| Conference | IEEE International Conference on Multimedia and Expo (ICME 2008) |
|---|---|
| City | Hannover |
| Period | 23/06/08 → 26/06/08 |
Keywords
- speech segmentation
- WEKA
- machine learning
- classifier
- LogitBoost