Stefan Werner

Effect of prosodic naturalness on segmental acceptability in synthetic speech

It is commonly agreed that one of the major goals in the development of modem text-to-speech synthesis is the improvement of prosody, especially intonation. Although high quality intonation is an important factor on the way to more natural synthetic speech, it is seldom scrutinized empirically whether and how this affects the relative performance of other components, such as segmental synthesis. The present paper discusses two preliminary rating experiments inquiring into the relation between the naturalness of intonation and subjective segmental quality in Finnish. Experiment 1 showed that the perception of intonation is dependent on the segmental quality. More crucially, experiment 2 indicated that also the perceived segmental acceptability is significantly dependent on the relative naturalness of intonation. In light of the present observations, the goal of improved intonation is not only desirable for the overall qualitys sake alone, but it is also shown to improve even the subjective perception of a very basic feature of synthetic speech such as segmental acceptability.

On the Role of Cognitive Abilities in Second Language Vowel Learning

This study investigated the role of different cognitive abilities—inhibitory control, attention control, phonological short-term memory (PSTM), and acoustic short-term memory (AM)—in second language (L2) vowel learning. The participants were 40 Azerbaijani learners of Standard Southern British English. Their perception of L2 vowels was tested through a perceptual discrimination task before and after five sessions of high-variability phonetic training. Inhibitory control was significantly correlated with gains from training in the discrimination of L2 vowel pairs. However, there were no significant correlations between attention control, AM, PSTM, and gains from training. These findings suggest the potential role of inhibitory control in L2 phonological learning. We suggest that inhibitory control facilitates the processing of L2 sounds by allowing learners to ignore the interfering information from L1 during training, leading to better L2 segmental learning.

Perceptual Training of Second-Language Vowels: Does Musical Ability Play a Role?.

The present study attempts to extend the research on the effects of phonetic training on the production and perception of second-language (L2) vowels. We also examined whether success in learning L2 vowels through high-variability intensive phonetic training is related to the learners’ general musical abilities. Forty Azerbaijani learners of Standard Southern British English participated in a pre-test/training/post-test setting. Discrimination and production tests were used in pre- and post-tests. The participants’ musical ability was evaluated through three musical aptitude tests. Results revealed a significant improvement in the discrimination and production of L2 vowels after training. However, the lack of a one-to-one relationship between production and perception gains suggests distinct representations underlying L2 speech perception and production. There was no significant correlation between overall musical ability scores and L2 vowel learning, however, results revealed a correlation between discrimination improvements and tonal memory. This suggests tonal memory facilitates the perceptual learning of the novel phonological structure of L2.

Autistic traits and attention to speech: Evidence from typically developing individuals

Abstract Individuals with autism spectrum disorder have a preference for attending to non-speech stimuli over speech stimuli. We are interested in whether non-speech preference is only a feature of diagnosed individuals, and whether we can we test implicit preference experimentally. In typically developed individuals, serial recall is disrupted more by speech stimuli than by non-speech stimuli. Since behaviour of individuals with autistic traits resembles that of individuals with autism, we have used serial recall to test whether autistic traits influence task performance during irrelevant speech sounds. The errors made on the serial recall task during speech or non-speech sounds were counted as a measure of speech or non-speech preference in relation to no sound condition. We replicated the serial order effect and found the speech to be more disruptive than the non-speech sounds, but were unable to find any associations between the autism quotient scores and the non-speech sounds. Our results may indicate a learnt behavioural response to speech sounds.

An acoustic description of spectral and temporal characteristics of Azerbaijani vowels

Abstract This paper presents an acoustic description of the nine Azerbaijani vowels; investigating the underlying acoustic and temporal characteristics of its vowel system. We explored acoustic and temporal parameters including: the first three formants (F1, F2 and F3), fundamental frequency (F0) and duration of the vowels. Participants in this study were 20 male and 23 female Azerbaijani speakers with a Tabrizi dialect. They were asked to utter three repetitions of the nine Azerbaijani vowels in three natural word contexts, embedded in carrier sentences. Results showed that the [ɯ] and [œ] vowels had a large overlap in the F1–F2 vowel space. Further analysis suggested that F3 is an important cue in discrimination of this vowel pair. Vowel-intrinsic duration effect seemed to be relatively strong in Azerbaijani. Other universal features also were found in the production of Azerbaijani vowels: low vowels and female speakers had lower F0 values. Surprisingly, in contrast with previous results for most languages, the average duration of Azerbaijani vowels was greater in males than females. The results of this study define the acoustic vowel-space of the Azerbaijani language and develop a database for further comparisons and investigations.

Vowel quantity and quality perception in Finnish and Estonian speakers.

Vowel quality perception in quantity languages might be expected to be unrelated to duration of the vowels since duration is used to realize phonemic duration oppositions. Our studies, both earlier reported ones and our latest project, indicate, though, that vowel‐intrinsic duration changes have a consistent effect on vowel identification in Estonian and Finnish speakers. Our latest experimental setup for investigating perceptional effects of microduration consists of two blocks, first a set of tests to identify the subject’s formant‐based categorial boundaries between vowel pairs on the close‐open axis, then the accordingly selected listening tests presenting formant‐wise ambiguous vowel exemplars produced with varying durations. Results from ten Estonian and ten Finnish speakers will be presented. In addition, the phenomenon observed will be compared with formant structure and duration perception in languages where these parameters co‐vary (e.g., English) and thus an influence of sub‐phonemic durational...

Close coupling or points of rendezvous? Connections between intonational events and the segmental grid

The authors examine the connection between the fundamental frequency (f0) contour and the segmental grid, employing the Fujisaki model for decomposing an f0 contour into phrase and accent components. As shown in earlier work by the first author, accent commands are closely connected with the accented syllables of the constituent words of an utterance. They reflect the temporal alignment of intonational gestures by their on‐ and offset times, as well as the magnitude of the gestures by their amplitudes. Their timing relates to phonemic function, and the amplitudes are strongly correlated with the prominence of an accented syllable. In this study the authors attempt to model the intensity contour of an utterance as a concomitant of the articulatory gesture, namely by low‐pass filtering pulse‐wise syllable commands, yielding onset and offset points as well as amplitude values of the syllabic gestures and relate these data to the accent command data from the f0 tier. We examine the influence of phonemic dis...