Prosodic boundary detection using syntactic and acoustic information. (January 2019)

• Record Type: Journal Article
• Authors: Kocharov, Daniil; Kachkovskaia, Tatiana; Skrelin, Pavel
• Is Part Of: Computer speech & language. Volume 53(2019)
• Page Start: 231
• Page End: 241
• Publication Date: 2019-01
• Subjects: Prosodic phrasing -- Automatic boundary detection -- Dependency parsing -- Acoustic feature -- Russian; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2018.07.001
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: Information on minor syntactic constituents significantly improves prosodic boundary detection based on acoustic measurements. Acoustic measurements taken over the stressed syllable give almost the same results as over the whole prosodic word. The hierarchy of acoustic cues signalling prosodic boundaries is speaker-dependent. Only half of the speakers use pause as the only cue for prosodic boundaries. Abstract: This paper presents a two-stage procedure for automatic prosodic boundary detection in Russian based on textual and acoustic data. The key idea of the method is (1) to predict all potential prosodic boundaries based on syntax and (2) among these potential boundaries, to choose those which are marked acoustically. For the first stage we developed a system which predicted a potential boundary whenever two adjacent words were not connected with each other in terms of syntax; for this we used a dependency tree parser and added several simple rules. At the second stage we run a random forest classifier to detect the actual prosodic boundaries using a small set of acoustic features. Of all the observed prosodic features pause duration worked best, and for some speakers it could be used as the only acoustic cue with no change in efficiency. For other speakers, however, other features were useful, such as tempo and amplitude resets or F0 range, and the choice of the features was speaker-dependent. In the end the procedure worked with the F1 measure of 0.91, recall of 0.90 and precision of 0.93, which is the best published result for Russian.