Does prosodic constituency signal relative predictability? A Smooth Signal Redundancy hypothesis

Abstract

This paper explores issues relating to signaling word boundaries from the perspec tive of Aylett's Smooth Signal Redundancy proposal (Aylett 2000, Aylett andTurk 2004) that language has evolved to spread redundancy, i.e. recognition likelihood, evenly throughout utterances. In Aylett's proposal, information that enables lis teners to identify sequences of elements in an utterance (signal redundancy) comes from two sources: a) language redundancy, recognition likelihood based on lex ical, syntactic, semantic, pragmatic and other factors, and b) acoustic redundancy, recognition likelihood based on acoustic salience. Smooth signal redundancy is achieved by a complementary relationship between language redundancy and acoustic redundancy that is implemented via prosodic structure.While Aylett and Turk (2004) present the case for prosodic prominence as a l ever for modulating the acoustic salience of syllables, the current paper proposes that prosodic constituency also fulfils this function for words. The current paper proposes that the signal redundancy, or recognition likelihood, of words can be manipulated by signaling their boundaries, and that the occurrence and strength of these boundary markers correlates inversely with language redundancy. Pro sodic constituency implements the complementary relationship between language redundancy and word boundary salience.Smooth Signal Redundancy provides an integrated explanation for a set of prop erties relating to prosodic constituent structure.

Dimensions of Segmental Variability: Interaction of Prosody and Surprisal in Six Languages

“…Our findings show that in accordance with Pellegrino et al (2011), tempo does not appear to interact with local segmental modulations of duration that encode information. These results might also speak in favor of Turk's suggestion (Turk, 2010) that tempo encodes acoustic redundancy globally, without interfering with local modulations or with the surprisal effects on, for instance, duration. Our data contains tempo conditions in which the speakers were reading the same text in three intended speech rates, normal, slow and fast.…”

“…For example, Bell et al (2009), Gahl et al (2012, Gahl (2015) and Cohen Priva (2015) also include prosodic effects such as speech rate or lexical stress as control factors, but crucially without attributing prosodic variation with a function that mediates, for instance, retrieval speed. Turk (2010) and Turk and Shattuck-Hufnagel (2014) also suggested a link between prosodic constituency and predictability. They proposed that signaling word boundary strength can be exploited to complement language redundancy.…”

“…The recognition likelihood of an element depends on linguistic and acoustic redundancy. Linguistic redundancy is defined as the lexical, syntactic, semantic and pragmatic contextual "clues to identity" of the element (Turk, 2010), or the element's predictability profile (Turk and ShattuckHufnagel, 2014). Acoustic redundancy, i.e., the set of acoustic clues to identity, is implemented by prosodic variation (Turk, 2010) or the prosodic profile.…”

“…Linguistic redundancy is defined as the lexical, syntactic, semantic and pragmatic contextual "clues to identity" of the element (Turk, 2010), or the element's predictability profile (Turk and ShattuckHufnagel, 2014). Acoustic redundancy, i.e., the set of acoustic clues to identity, is implemented by prosodic variation (Turk, 2010) or the prosodic profile. The inverse relationship between predictability and the prosodic profile results in a Smooth Signal Redundancy (SSR) profile, that is, in the smooth spread of information over the utterance, thereby globally optimizing the recognition likelihood of the utterance.…”

“…We also aim to shed more light on the research questions posed by Turk (2010), Turk and Shattuck-Hufnagel (2014), and Pellegrino et al (2011) regarding the relationship between speech rate and surprisal. Specifically, we address the hypothesis that the effect of surprisal does not depend on speech rate but is robustly positive across speech rate levels in all studied languages.…”