César González Ferreras

From HTML to VoiceXML: A First Approach

In this work, we discuss the construction process of the voice portal counterpart of a departmental web site. VoiceXML has been used as the dialogue modelling language. A prototypical system has been built using our own VoiceXML interpreter, which easily integrates different implementation platforms. A general discussion of VoiceXML advantages and disadvantages is reported and a simple startup procedure is proposed as a means to build voice portals starting from legacy web sites.

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

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.

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

In this paper we present an experimental study on how corpus-based automatic prosodic information labeling can be transferred from a source language to a different target language. The Spanish ESMA corpus is used to train models for the identification of the prominent words. Then, the models are used to identify the accented words of the English Boston University Radio News Corpus (BURNC). The inverse process (training the models with English data and testing with the Spanish corpus) is also contrasted with the results obtained in the conventional scenario: training and testing using the same corpus. We got up to 82.7% correct annotation rates in cross-lingual experiments, which contrast slightly with the accuracy obtained in a mono-lingual single speaker scenarios (86.6% for Spanish and 80.5% for English). Speaker independent monolingual recognition experiments have been also performed with the BURNC corpus, leading to cross-speakers results that go from 69.3% to 84.2% recognition rates. As these results are comparable to the ones obtained in the cross-lingual scenario we conclude that the new approach we defend has to face up with similar challenges as the ones presented in speaker independent scenarios.

Evaluating Different Non-native Pronunciation Scoring Metrics with the Japanese Speakers of the SAMPLE Corpus

This work presents an analysis over the set of results derived from the goodness of pronunciation (GOP) algorithm for the evaluation of pronunciation at phoneme level over the SAMPLE corpus of non native speech. This corpus includes several recordings of uttered sentences by distinct speakers that have been rated in terms of quality by a group of linguists. The utterances have been automatically rated with the GOP algorithm. The phoneme dependence is discussed to suggest the normalization of intermediate results that could enhance the metrics performance. As result, new scoring proposals are presented which are based on computing the log-likelihood values obtained from the GOP algorithm and the application of a set of rules. These new scores show to correlate with the human rates better than the original GOP metric.