Analysis of inconsistencies in cross-lingual automatic ToBI tonal accent labeling

Authors:
David Escudero-Mancebo;Carlos Vivaracho Pascual;César González Ferreras;Valentín Cardeñoso-Payo;Lourdes Aguilar
Affiliations:
Dpt. of Computer Sciences, Universidad de Valladolid, Spain;Dpt. of Computer Sciences, Universidad de Valladolid, Spain;Dpt. of Computer Sciences, Universidad de Valladolid, Spain;Dpt. of Computer Sciences, Universidad de Valladolid, Spain;Dpt. of Spanish Philology, Universidad Autónoma de Barcelona, Spain
Venue:
TSD'11 Proceedings of the 14th international conference on Text, speech and dialogue
Year:
2011

Citing 7
Cited 2

Data mining: practical machine learning tools and techniques with Java implementations

Data mining: practical machine learning tools and techniques with Java implementations
Automatic ToBI prediction and alignment to speed manual labeling of prosody

Speech Communication - Special issue on speech annotation and corpus tools
SMOTE: synthetic minority over-sampling technique

Journal of Artificial Intelligence Research
POST: using probabilities in language processing

IJCAI'91 Proceedings of the 12th international joint conference on Artificial intelligence - Volume 2
The WEKA data mining software: an update

ACM SIGKDD Explorations Newsletter
Automatic Prosodic Event Detection Using Acoustic, Lexical, and Syntactic Evidence

IEEE Transactions on Audio, Speech, and Language Processing
Exploiting Acoustic and Syntactic Features for Automatic Prosody Labeling in a Maximum Entropy Framework

IEEE Transactions on Audio, Speech, and Language Processing

Cross-lingual English Spanish tonal accent labeling using decision trees and neural networks

NOLISP'11 Proceedings of the 5th international conference on Advances in nonlinear speech processing
Glissando: a corpus for multidisciplinary prosodic studies in Spanish and Catalan

Language Resources and Evaluation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an experimental study on how corpus-based automatic prosodic information labeling can be transferred from a source language to a different target language. Tone accent identification models trained for Spanish, using the ESMA corpus, are used to automatically assign tonal accent ToBI labels on the (English) Boston Radio news corpus, and vice versa. Using just local raw prosodic acoustic features, we got about 75% correct annotation rates, which provides a good starting point to speed up automatic prosodic labeling of new unlabeled corpora. Despite the different ranges and relevance of inter corpora acoustic input features, the contrasting of the results with respect to manual labeling profiles indicate the potential capabilities of the procedure.