Classification of Filipino Speech Rhythm Using Computational and Perceptual Approach

Classification of Filipino Speech Rhythm Using Computational and Perceptual Approach Timothy Israel D. Santos a, Rowena Cristina L. Guevara Ph.D. b Digital Signal Processing Laboratory, Electrical and Electronics Engineering Institute, University of the Philippines, Diliman atdsantos1@up.edu.ph bgev@eee.upd.edu.ph Abstract. This study incorporates computational and perceptual methods to classify Filipino speech rhythm. Speech rhythm may be described as a language‟s distinguishing durational sound pattern, resulting from the complexity of the language‟s syllable inventory. 1 Computational methods involve the correlation of rhythm-types to acoustic features such as the vocalic and consonantal intervals, one of which is the implementation of Multivariate Discriminant Analysis (MDA). Perceptual methods involve contrasting the rhythm of an unclassified language from prototype syllable-timed and stress-timed sentences. In order to isolate rhythm from speech, a data-stripping technique called flat sasasa resynthesis was implemented wherein the consonants are replaced with /s/ and vowels with /a/, producing a resynthesized alternating “sasasa” sounds at a constant pitch (F0). The rhythm discrimination and classification were closely examined for consistency between the data modeling and listening test results. The computational experiment was able to show that an MDA classifier trained to distinguish English and Japanese sentences tend to label Filipino sentences as Japanese 67% of the time, vis-à-vis the perceptual experiment showing that the listeners perceive Filipino to be more similar with Japanese, this study shows computational and perceptual validation that Filipino is syllable-timed, just like Japanese. Keywords: Computational Linguistics, Natural Language Processing, Psychoacoustics, Speech Analysis, Speech Rhythm 1 Introduction Speech prosody is an important aspect of natural speech. Pitch, duration, and intensity are some of the prosodic components that dictate the naturalness of speech and differentiate contextual meaning. The objective of this study is to present computational analyses and perceptual experiments to confirm the existence and applicability of rhythm classes to Filipino; to find their salient features for computational classification; and to find the rhythm typology of Filipino speech. Among the prosodic characteristics of speech, this study focuses on the temporal or the duration aspect. Speech prosody or suprasegmentals refer to the features of speech that can modify the meaning and information being carried by the words not found in the lexicon. Speech intensity, duration, pitch, and rhythm are the most common features that differentiate one prosodic unit from another. Embedded in the human perception is the distinction of speech rhythm or timing to discriminate between languages. Languages are traditionally classified as stress-timed or syllable-timed where syllable-timed language were said to exhibit near-equal duration for each syllable, and stress-timed languages were said to have near-equal duration between stresses or 1 Some languages allow a variety of complex syllables CnV, CVC, etc., while some languages strictly adhere to simpler CV syllables (C-consonant, V-vowel). 25th Pacific Asia Conference on Language, Information and Computation, pages 450–458 450 what is called isochrony (Pike, 1943). Years of research on the subject has shown development in which evidences have been piling up against isochrony theory, and there emerged other rhythm classes such as Mora-timing (Abercrombie, 1967; Benadon, 2009). New models have …

Content truncated to 500 words