Modeling tones in continuous Cantonese speech
Cantonese is a major Chinese dialect with a complicated tone system. This research focuses on quantitative modeling of Cantonese tones. It uses Stem-ML, a language-independent framework for quantitative intonation modeling and generation. A set of F0 prediction models are built, and trained on acous...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal article |
| Language: | English |
| Published: |
2002
|
| Subjects: |
| _version_ | 1826306611921027072 |
|---|---|
| author | Lee, T Kochanski, G Shih, C Li, Y |
| author2 | International Speech Communication Association (ISCA) |
| author_facet | International Speech Communication Association (ISCA) Lee, T Kochanski, G Shih, C Li, Y |
| author_sort | Lee, T |
| collection | OXFORD |
| description | Cantonese is a major Chinese dialect with a complicated tone system. This research focuses on quantitative modeling of Cantonese tones. It uses Stem-ML, a language-independent framework for quantitative intonation modeling and generation. A set of F0 prediction models are built, and trained on acoustic data. The prediction error is about 11 Hz of 1 semitone. The resulting optimal model parameters are analyzed in accordance with linguistic knowledge. Key observations include: (1) There is no obvious advantage to model the entering tones separately. They can be considered as simply truncated versions of the non-entering tones; (2) Cantonese appears to have a declining phrase intonation; (3) Tones at initial positions of a phrase or a sentence tend to have a greater prosodic strength than those at the final positions; (4) Content words are stronger than function words; (5) Long words are stronger than short words. |
| first_indexed | 2024-03-07T06:50:35Z |
| format | Journal article |
| id | oxford-uuid:fc6d8ab1-4301-4bc8-99f0-803031cf1801 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T06:50:35Z |
| publishDate | 2002 |
| record_format | dspace |
| spelling | oxford-uuid:fc6d8ab1-4301-4bc8-99f0-803031cf18012022-03-27T13:20:42ZModeling tones in continuous Cantonese speechJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:fc6d8ab1-4301-4bc8-99f0-803031cf1801LinguisticsEnglishOxford University Research Archive - Valet2002Lee, TKochanski, GShih, CLi, YInternational Speech Communication Association (ISCA)Cantonese is a major Chinese dialect with a complicated tone system. This research focuses on quantitative modeling of Cantonese tones. It uses Stem-ML, a language-independent framework for quantitative intonation modeling and generation. A set of F0 prediction models are built, and trained on acoustic data. The prediction error is about 11 Hz of 1 semitone. The resulting optimal model parameters are analyzed in accordance with linguistic knowledge. Key observations include: (1) There is no obvious advantage to model the entering tones separately. They can be considered as simply truncated versions of the non-entering tones; (2) Cantonese appears to have a declining phrase intonation; (3) Tones at initial positions of a phrase or a sentence tend to have a greater prosodic strength than those at the final positions; (4) Content words are stronger than function words; (5) Long words are stronger than short words. |
| spellingShingle | Linguistics Lee, T Kochanski, G Shih, C Li, Y Modeling tones in continuous Cantonese speech |
| title | Modeling tones in continuous Cantonese speech |
| title_full | Modeling tones in continuous Cantonese speech |
| title_fullStr | Modeling tones in continuous Cantonese speech |
| title_full_unstemmed | Modeling tones in continuous Cantonese speech |
| title_short | Modeling tones in continuous Cantonese speech |
| title_sort | modeling tones in continuous cantonese speech |
| topic | Linguistics |
| work_keys_str_mv | AT leet modelingtonesincontinuouscantonesespeech AT kochanskig modelingtonesincontinuouscantonesespeech AT shihc modelingtonesincontinuouscantonesespeech AT liy modelingtonesincontinuouscantonesespeech |