Exploring the Nuances of Reduction in Conversational Speech: Lexicalized and Non Lexicalized Reductions Kübra (original) (raw)
Related papers
An Acoustic Profile Of Consonant Reduction
1996
Vowel reduction has been studied for years. It is a universal phenomenon that reduces the distinction of vowels in informal speech and unstressed syllables. How consonants behave in situations where vowels are reduced is much less well known. In this paper we compare durational and spectral data (for both intervocalic consonants and vowels) segmented from read speech with otherwise identical segments from spontaneous speech. On a global level, it shows that consonants reduce like vowels when the speaking style becomes informal. On a more detailed level there are differences related to the type of the consonant.
Phonetic reductions and linguistic factors
2013
In natural communication it is common for speakers to vary between distinct and reduced pronunciations of words or phonemic strings. This paper highlights the some results from a recent large scale study of the occurence of phonetic reductions in Danish spontaneous speech. In this study phonetic reduction is explored by mapping the abstract phonemic representation in a spontaneous speech corpus with the actual phonetic realization on a phone-by-phone basis. By investigating the occurence of distinct vs. reduced realizations of phonemes, it is demonstrated that the propensity for phonetic reduction is closely related to various levels of linguistic description, e.g. the articulatory traits of the individual phonemes, their phonological context, morphological structure, grammatical function and pragmatic factors.
Why reduce? Phonological neighborhood density and phonetic reduction in spontaneous speech
Journal of Memory and Language, 2012
Frequent or contextually predictable words are often phonetically reduced, i.e. shortened and produced with articulatory undershoot. Explanations for phonetic reduction of predictable forms tend to take one of two approaches: Intelligibility-based accounts hold that talkers maximize intelligibility of words that might otherwise be difficult to recognize; production-based accounts hold that variation reflects the speed of lexical access and retrieval in the language production system. Here we examine phonetic variation as a function of phonological neighborhood density, capitalizing on the fact that words from dense phonological neighborhoods tend to be relatively difficult to recognize, yet easy to produce. We show that words with many phonological neighbors tend to be phonetically reduced (shortened in duration and produced with more centralized vowels) in connected speech, when other predictors of phonetic variation are brought under statistical control. We argue that our findings are consistent with the predictions of production-based accounts of pronunciation variation.
Journal of Phonetics, 2011
In spontaneous, conversational speech, words are often reduced compared to their citation forms, such that a word like yesterday may sound like ['jESeI]. The present paper investigates such acoustic reduction . The study of reduction needs large corpora that are transcribed phonetically. The first part of this paper describes an automatic transcription procedure used to obtain such a large phonetically transcribed corpus of Dutch spontaneous dialogues, which is subsequently used for the investigation of acoustic reduction. First, the orthographic transcription were adapted for automatic processing. Next, the phonetic transcription of the corpus was created by means of a forced alignment with a lexicon with multiple pronunciation variants per word. These variants were generated by applying phonological and reduction rules to the canonical phonetic transcriptions of the words. The second part of this paper reports the results of a quantitative analysis of reduction in the corpus on the basis of the generated transcriptions and gives an inventory of segmental reductions in standard Dutch. Overall, we found that reduction is more pervasive in spontaneous Dutch than previously documented.
Lexical competition and reduction in speech: A preliminary report
Research on Spoken Language Processing Progress …, 1997
among others has proposed that talkers accommodate listeners' communicative needs by controlling the degree of reduction (hyper-and hypo-articulation) in different contextual conditions, thereby maintaining sufficient intelligibility of words across a variety of contexts. Lindblom's proposal predicts that lexical factors that affect intelligibility of a word will affect the hypo-and hyperarticulation of words. Based on factors in lexical competition such as usage frequency and similarity-neighborhood density, previous research has characterized words as "easy" or "hard" to identify. This study examines the degree of centralization of vowels (a well known feature of reduction or hypo-articulation) in 34 "easy" and 34 "hard" monosyllabic (CVC) words of equal familiarity spoken in isolation by 10 talkers. Measurements of the first two vowel formants (F1, F2) were made at the point of maximal displacement in the vowel (excluding the initial and final 50 ms of the vowel). Centralization is measured by calculating the Euclidean distance from the center of a talker's F1-F2 vowel space. Three results emerge: 1) overall "easy" words were significantly more centralized than "hard" words, 2) peripheral vowels, such as /i/, /a/, /u/, showed the greatest effect, and 3) there was considerable between talker variability in the magnitude of the difference between vowels in "easy" and "hard" words. The results are interesting because they demonstrate that the talker takes into account a wider variety of sources of possible noise and information than previously thought. These results have implications for both diachronic and synchronic processes that involve reduction in linguistics. They have further implications for applications in speech recognition that model variation in spoken language.
An acoustic description of consonant reduction
Speech Communication, 1999
The acoustic consequences of the articulatory reduction of consonants remain largely unknown. Much more is known about acoustic vowel reduction. Whether the acoustical and perceptual consequences of articulatory consonant reduction are comparable in kind and extent to the consequences of vowel reduction is still an open question. In this study we compare acoustic data for 791 VCV realizations, containing 17 Dutch intervocalic consonants and 13 vowels, extracted from read speech from a single male speaker, to otherwise identical segments isolated from spontaneous speech. Five acoustic correlates of reduction were studied. Acoustic tracers of articulation were based on F 2 slope dierences and locus equations. Speech eort was assessed by measuring duration, spectral balance, and the intervocalic sound energy dierence of consonants. On a global level, it shows that consonants reduce acoustically like vowels on all investigated accounts when the speaking style becomes informal or syllables become unstressed. Methods that are sensitive to speech eort proved to be more reliable indicators of reduction than F 2 based measures. On a more detailed level there are dierences related to the type of consonant. The acoustic results suggest that articulatory reduction will decrease the intelligibility of consonants and vowels in comparable ways.
Laboratory Phonology, 2012
In spontaneous speech, words may be realised shorter than in formal speech (e.g., English yesterday may be pronounced like [jɛ∫eı]). Previous research has shown that context is required to understand highly reduced pronunciation variants. We investigated the extent to which listeners can predict low predictability reduced words on the basis of the semantic/syntactic and acoustic cues in the context. In four experiments, participants were presented with either the preceding context or the preceding and following context of reduced words, and either heard these fragments of conversational speech, or read their orthographic transcriptions. Participants were asked to predict this missing reduced word on the basis of the context alone, choosing from four plausible options.
The study aims to compare vowel reduction in read and fully spontaneous speech in English and Polish. It hypothesizes that (i) vowels exhibit stronger reduction in fully spontaneous speech in comparison with read speech in the two languages (ii) vowel reduction is more robust in English than it is in Polish (iii) high speech rate of triggers vowel reduction. The aims were achieved by an acoustic analysis of interviews and word lists from PAC (9 speakers) and the Corpus of Modern Spoken Polish in the area of Greater Poland (9 speakers). The study takes centralization of formants and short vowel duration as vowel reduction (Lindblom 1963) which were normalized to compare the values across speakers. For Polish subjects, speakers’ canonical schwa was operationalized as an average of peripheral vowels /i/, /a/ and /u/ due to the fact that Polish has no schwa (Jassem 2003). Comparison of two speech styles consisted in measuring spectral and temporal properties of vowel tokens from the wordlist and from interviews. The rate-reduction hypothesis was tested by means of comparing vowel reduction for three fastest and three slowest speakers for each language and using Pearson correlation. In light of the obtained results, the two first hypotheses were positively verified. The third one produced mixed results. The study establishes a significant difference in vowel reduction across two speech styles, read and fully spontaneous across two unrelated languages. All vowel tokens were shorter and centralized in spontaneous speech, relative to their duration as well as placed in less peripheral positions than in read speech. It has been shown that reduction in English is considerably stronger than in Polish. With respect to the third hypothesis, assuming a straightforward relationship between speech rate and reduction, the findings of the current study did not provide a definite answer. To a certain extent, the correlation between rate and duration was found in Polish but not in English. As Zwicky notes, “casual speech need not to be fast; some speakers [...] use a quite informal speech even at fairly slow rates of speech, while others [...] give the impression of great precision even in hurried speech” (Zwicky 1972: 607).
Modelling Aspects of Reduction and Assimilation of Consonant Sequences in Spontaneous French Speech
International Workshop on Structural and Syntactic Pattern Recognition, 2007
The following paper first presents spectrographic data of consonant sequences containing one or two consonants omitted and/or changed into another consonant when compared to a previous perception analysis. In most cases, perceptual and acoustic data are seen to strongly correspond, proving that consonants had indeed be changed, significantly reduced or deleted, mainly in a weak position, thereby preserving acoustic information