A Multi-Level Context-Dependent Prosodic Model
Applied to Durational Modeling

Nicolas Obin (1) , Xavier Rodet (1) , Anne Lacheret (2)

1 STMS - Sciences et Technologies de la Musique et du Son
2 MoDyCo - Modèles, Dynamiques, Corpus

We present in this article a multi-level prosodic model based on the estimation of prosodic parameters on a set of well defined linguistic units. Different linguistic units are used to represent different scales of prosodic variations (local and global forms) and thus to estimate the linguistic factors that can explain the variations of prosodic parameters independently on each level. This model is applied to the modeling of syllable-based durational parameters on two read speech corpora - laboratory and acted speech. Compared to a syllable-based baseline model, the proposed approach improves performance in terms of the temporal organization of the predicted durations (correlation score) and reduces model's complexity, when showing comparable performance in terms of relative prediction error.

Keywords: speech synthesis; prosody; multi-level model; context-dependent model

Domains: Linguistics; Machine Learning [stat.ML]; Signal and Image processing; Signal and Image Processing

Nicolas Obin, Xavier Rodet, Anne Lacheret. A Multi-Level Context-Dependent Prosodic Model
Applied to Durational Modeling. Interspeech, Sep 2009, Brighton, United Kingdom. ⟨halshs-00419976⟩