This study compared segmental distribution patterns for consonants and vowels in English infant-directed speech (IDS) and adult-directed speech (ADS). A previous study of Korean indicated that segmental patterns of IDS differed from ADS patterns (Lee, Davis & MacNeilage, 2008). The aim of the current study was to determine whether such differences in Korean are universal or language-specific. R...

This study investigates how similarly present and absent English phonemes behind noise are perceived by native and non-native speakers. Participants were English native speakers and Japanese native speakers who spoke English as a second language. They listened to English words and non-words in which a phoneme was covered by noise (added; phoneme + noise) or replaced by noise (replaced; noise on...

The transmission protocol of sustained voiced speech is hypothesized to be based on a fundamental drive process, which synchronizes the vocal tract excitation on the transmitter side and evokes the pitch perception on the receiver side. A band limited fundamental drive is extracted from a voice specific subband decomposition of a speech signal. When the near periodic drive is used as fundamenta...

This paper presents and evaluates a novel text-directed speech enhancement algorithm for usage in non real-time applications. In our approach, the text of the intended dialogue is used to partition noisy speech into regions of broad phoneme classiications. Classes considered include stops, fricatives, aaricates, nasals, vowels, semivowels, diphthongs and silence. These partitions are then used ...

Totally 41 postlingually deafened adult patients who have received multi-channel cochlear implants at Stockholm Söder Hospital have participated in this study. All patients have followed the CI-programme at the clinic and have been tested 0, 6, 12, and 24 months postoperatively. The test-battery consisted of vowels (cVc) and consonants (vCv), Everyday Sentences(ES), and Connected Discourse Trac...

In order to make examples morphologically transparent, we have transcribed certain vowels separately between morpheme boundaries even though some of these vowels are elided or coalesced in speech. Our segmental transcriptions follow Mugane 1996 and 1997. The seven vowels transcribed i, ı̃, e, a, o ũ, u have the approximate respective values [i, e, ε, a, , o, u], each of which can be short or lon...

Over the past decade, computational linguists have been striving to design a speech recognition system that is able to identify standard speeches and to accommodate sound variables caused by different individual accents. Furthermore, some speech recognition programs have been able to learn and identify distinctive sound frequencies due to the user’s age and gender. Nevertheless, regular sound a...

An original three-dimensional (3D) linear articulatory model of the velum and nasopharyngeal wall has been developed from magnetic resonance imaging (MRI) and computed tomography images of a French subject sustaining a set of 46 articulations, covering his articulatory repertoire. The velum and nasopharyngeal wall are represented by generic surface triangular meshes fitted to the 3D contours ex...

It is assumed linguistic symbols must be grounded in perceptual information to attain meaning, because the sound of a word in a language has an arbitrary relation with its referent. This paper demonstrates that a strong arbitrariness claim should be reconsidered. In a computational study, we showed that one phonological feature (nasals in the beginning of a word) predicted negative valence in t...

In our search for perceptually relevant cues in speech we have taken special interest in pa r t s of the acoustic signal displaying rapid changes and short durations. In "tape splicing" e-xperiments with natural voiceless stops we have found a greater amount of absolute identity of the esplosion than might be expected from ear l ie r studies. This finding ~hou.ld. be taken into accou;lt in spee...