Native phonotactics influences speech perception, as numerous studies have shown. The present study tackles the question whether there is a difference in perceptual performance if the involved sequence occurs only in loanwords, compared to a sequence that does not occur at all in the native language. This was tested with the native Japanese sequences of palatal affricate plus /i/, compared to /ti/ (accepted only in loanwords) versus /zi/ (not accepted in Japanese) in an online AX discrimination task with 39 Japanese speakers (21-63 years old), who also had to answer three questions on their received English input. Participants performed significantly better at discriminating the accepted loan sequence /ti/, though discrimination of the foreign sequence /zi/ was also quite high (ranging from 40-100% correct). The results indicate that discriminability is only partly guided by native phonotactics. A potential role of amount of English input measured by self- report could not be attested.

This study examines whether creaky voice serves as a perceptual cue for the fortis-lenis word-initial stop distinction in Standard Seoul Korean. Twenty-nine native speakers com- pleted an online forced-choice ABX task, where they deter- mined whether a manipulated, artificially creaky lenis token (Sound X) with ambiguous values of VOT and F0 resembled more closely a natural fortis (Sound A) or a natural lenis to- ken (Sound B). Results showed considerable inter-speaker vari- ation: some rarely categorized creaky lenis tokens as fortis, while others did so up to 15% of the time. Statistical analysis showed that presence of creak in Sound X did not change the perceptual classification significantly, suggesting creak is not a primary cue alongside VOT and F0. Word-specific effects also emerged, highlighting the complexity of perceptual cues in Ko- rean stop contrasts.

We present an exploratory analysis of F0 ratio, a proposed method to pick up octave jumps in the speech signal. Such jumps, often considered errors, are possibly indicative of the presence of creaky voice. This paper focuses on co-intrinsic voice quality in Seoul Korean fortis stops, building on previous data, and on the Danish contrastive voice quality stød. The re-sults suggest that in Seoul Korean F0 ratio captures a clear jump upwards indicating a modality switch from creaky to modal voice, with a gender difference observed. This suggests that pitch jumps are not necessarily erroneous but may reflect sys-tematic cues to phonological contrasts cued with creak. In Dan-ish, F0 ratio captures a rising intonational contour for non-stød tokens and is able to categorise between stød and non-stød to-kens with high accuracy, although this leaves open the question whether F0 ratio captures modality shifts in Danish, or rather a combination of modality shift and pitch contour differences.

Pitch tracking algorithms can show upward or downward jumps in F0 by one octave. These “octave jumps” are sometimes thought of as pitch-tracking “errors”, in the sense that they constitute a “mistake” in the algorithm. Using Praat software, we discuss the point (which has been made before) that measured octave jumps often actually reflect genuine changes in periodicity and glottal-fold vibration. We illustrate this with the example of creaky voice in fortis stops in Seoul Korean. We argue (1) that when the goal is to capture periodicity or vocal-fold vibration, pitch-tracking algorithms capture F0 well, with pitch jumps possibly reflecting an important language-specific feature, and (2) that ignoring such jumps (due to assuming an error) could lead to misrepresentation of the properties of the language. To quantify these real F0 jumps, we introduce the notion of the “F0 ratio”, which identifies potential F0 jumps and helps to chart the frequency of pitch jumps in a language.

This study investigates the stop voicing contrast in Palestinian Arabic (PA) by examining Voice Onset Time (VOT) in both production and perception. An acoustic analysis of the recordings of 8 speakers showed that word-initial voiced stops in sentence context have an average VOT of -93 msec, and word-initial voiceless stops one of 29 msec. PA thus belongs, like most dialects of Arabic, to true voicing languages, i.e., languages with a contrast between voicing lead and short lag VOT. We furthermore tested whether the phoneme /b/, without voiceless counterpart /p/ in PA, has similar VOT values to /d, dʕ/, which have voiceless counterparts /t, tʕ/. Similarly, we compared /k/, without counterpart /g/ in the PA dialect we investigated, to /t, tʕ/. For /b/ we found very similar VOT values to /d, dʕ/, while for /k/ we found a difference to /t, tʕ/, attributable to a general tendency of velars to have longer VOT than denti- alveolars. We thus found no evidence for a less contrastive realization of unpaired plosives in PA. In a categorization experiment of the denti-alveolar phoneme pairs with the same 8 speakers, VOT proved sufficient as a perceptual cue, though f0 of the following vowel also influenced the categorization.

Prior research has revealed that L1-Mandarin learners employed position-dependent repair strategies for European Portuguese /l/ and /ɾ/. In this study we examined whether this L2 prosodic effect can be attributed to a cross-linguistic influence and whether the replacement of the Portuguese rhotic by the Mandarin [ɻ] is due to perception or orthography. We performed a delayed imitation task with naïve Mandarin listeners and manipulated the presented input types (auditory form alone or a combination of auditory and written forms). Results showed that naïve responses were reminiscent of L1- Mandarin learners’ behaviour, and that [ɻ] was used almost exclusively in the presence of written input, suggesting that the prosodic effect attested in L2 acquisition of European Portuguese /l/ and /ɾ/ stems from cross-linguistic interaction between phonological categorization and orthography.

Previous observations from didactic studies have indicated that Vietnamese learners of German as a foreign language often fail to realize consonantal clusters in German [1, 2, 3]. The present study investigated whether this problem occurs already at the level of perception, i.e., whether Vietnamese learners find it difficult to perceive the difference between a cluster and a single consonant. We focused on the discrimination between the German cluster /ʃt/ and the single consonants /t/ and /ʃ/, both in onset and coda position. Due to different phonotactic restrictions on coda consonants in Vietnamese, we expected the coda position to pose a bigger challenge for correct discrimination than the onset position. With an AX discrimination task, we tested how 83 university students from Hanoi perceived these contrasts. Our findings show that only the distinction between /ʃt/-/ʃ/ in coda position posed a real challenge to our listeners. We attribute this difficulty to the weak and non-native auditory cues for the plosive in this position. For all other contrasts our participants performed surprisingly well. We propose that this is due to the influence of English as first L2 that facilitates the acquisition of phonological contrasts in German as an L3.