Using prosody to classify discourse relations
This work aims to explore the correlation between the discourse structure of a spoken monologue and its prosody by predicting discourse relations from different prosodic attributes. For this purpose, a corpus of semi-spontaneous monologues in English has been automatically annotated according to the...
Guardado en:
| Publicado: |
2017
|
|---|---|
| Materias: | |
| Acceso en línea: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_2308457X_v2017-August_n_p3201_Kleinhans http://hdl.handle.net/20.500.12110/paper_2308457X_v2017-August_n_p3201_Kleinhans |
| Aporte de: |
| id |
paper:paper_2308457X_v2017-August_n_p3201_Kleinhans |
|---|---|
| record_format |
dspace |
| spelling |
paper:paper_2308457X_v2017-August_n_p3201_Kleinhans2023-06-08T16:35:33Z Using prosody to classify discourse relations Discourse structure Prosody RST Speech synthesis Support vector machines Continuous speech recognition Speech Speech synthesis Support vector machines Text processing Discourse structure Prosodic features Prosody Rhetorical relations Rhetorical structure theory Speech rates Speech understanding Supervised classification Speech communication This work aims to explore the correlation between the discourse structure of a spoken monologue and its prosody by predicting discourse relations from different prosodic attributes. For this purpose, a corpus of semi-spontaneous monologues in English has been automatically annotated according to the Rhetorical Structure Theory, which models coherence in text via rhetorical relations. From corresponding audio files, prosodic features such as pitch, intensity, and speech rate have been extracted from different contexts of a relation. Supervised classification tasks using Support Vector Machines have been performed to find relationships between prosodic features and rhetorical relations.Preliminary results show that intensity combined with other features extracted from intra- and intersegmental environments is the feature with the highest predictability for a discourse relation. The prediction of rhetorical relations from prosodic features and their combinations is straightforwardly applicable to several tasks such as speech understanding or generation. Moreover, the knowledge of how rhetorical relations should be marked in terms of prosody will serve as a basis to improve speech synthesis applications and make voices sound more natural and expressive. Copyright © 2017 ISCA. 2017 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_2308457X_v2017-August_n_p3201_Kleinhans http://hdl.handle.net/20.500.12110/paper_2308457X_v2017-August_n_p3201_Kleinhans |
| institution |
Universidad de Buenos Aires |
| institution_str |
I-28 |
| repository_str |
R-134 |
| collection |
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) |
| topic |
Discourse structure Prosody RST Speech synthesis Support vector machines Continuous speech recognition Speech Speech synthesis Support vector machines Text processing Discourse structure Prosodic features Prosody Rhetorical relations Rhetorical structure theory Speech rates Speech understanding Supervised classification Speech communication |
| spellingShingle |
Discourse structure Prosody RST Speech synthesis Support vector machines Continuous speech recognition Speech Speech synthesis Support vector machines Text processing Discourse structure Prosodic features Prosody Rhetorical relations Rhetorical structure theory Speech rates Speech understanding Supervised classification Speech communication Using prosody to classify discourse relations |
| topic_facet |
Discourse structure Prosody RST Speech synthesis Support vector machines Continuous speech recognition Speech Speech synthesis Support vector machines Text processing Discourse structure Prosodic features Prosody Rhetorical relations Rhetorical structure theory Speech rates Speech understanding Supervised classification Speech communication |
| description |
This work aims to explore the correlation between the discourse structure of a spoken monologue and its prosody by predicting discourse relations from different prosodic attributes. For this purpose, a corpus of semi-spontaneous monologues in English has been automatically annotated according to the Rhetorical Structure Theory, which models coherence in text via rhetorical relations. From corresponding audio files, prosodic features such as pitch, intensity, and speech rate have been extracted from different contexts of a relation. Supervised classification tasks using Support Vector Machines have been performed to find relationships between prosodic features and rhetorical relations.Preliminary results show that intensity combined with other features extracted from intra- and intersegmental environments is the feature with the highest predictability for a discourse relation. The prediction of rhetorical relations from prosodic features and their combinations is straightforwardly applicable to several tasks such as speech understanding or generation. Moreover, the knowledge of how rhetorical relations should be marked in terms of prosody will serve as a basis to improve speech synthesis applications and make voices sound more natural and expressive. Copyright © 2017 ISCA. |
| title |
Using prosody to classify discourse relations |
| title_short |
Using prosody to classify discourse relations |
| title_full |
Using prosody to classify discourse relations |
| title_fullStr |
Using prosody to classify discourse relations |
| title_full_unstemmed |
Using prosody to classify discourse relations |
| title_sort |
using prosody to classify discourse relations |
| publishDate |
2017 |
| url |
https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_2308457X_v2017-August_n_p3201_Kleinhans http://hdl.handle.net/20.500.12110/paper_2308457X_v2017-August_n_p3201_Kleinhans |
| _version_ |
1768542150445760512 |