Chad Vicenik

VOICESAUCE: A program for voice analysis.

VOICESAUCE is a new application, implemented in MATLAB, which provides automated voice measurements over time from audio recordings. The measures currently computed are F0, H1(*), H2(*), H4(*), H1(*)‐H2(*), H2(*)‐H4(*), H1(*)‐A1, H1(*)‐A2, H1(*)‐A3, energy, Cepstral Peak Prominence, F1–F4, and B1–B4, where (*) indicates that harmonic amplitudes are reported with and without corrections for formant frequencies and bandwidths [Iseli et al. (2006)]. Formant values are calculated using the Snack Sound Toolkit, while F0 is calculated using the STRAIGHT algorithm; harmonic spectra magnitudes are computed pitch‐synchronously. VOICESAUCE takes as input a folder of wav files, and for each input wav file produces a MATLAB file with values every millsecond for all measures. It can operate over the whole input file or over segments delimited by a PRAAT textgrid file. VOICESAUCE then takes these MATLAB outputs, optionally along with electroglottographic measurements obtained separately from PCQUIRERX, and provides con...

An acoustic study of Georgian stop consonants

This study investigates the acoustic properties of ejective, voiced and voiceless aspirated stops in Georgian, a Caucasian language, and seeks to answer two questions: (i) Which acoustic features discriminate the three stop types? and (ii) Do Georgian stops undergo initial strengthening, and if so, is it syntagmatic or paradigmatic strengthening? Five female speakers were recorded reading words embedded in carrier phrases and stories. Acoustic measures include closure duration, voicing during the closure, voicing lag, relative burst intensity, spectral moment of bursts, phonation (H1-H2) and F0. Of these, voicing lag, voicing during the closure, mean burst frequency, H1-H2 and F0 could all be used to discriminate stop type, but stop types did not differ in closure duration or relative burst intensity. Georgian stops did show initial strengthening and showed only syntagmatic enhancement, not paradigmatic enhancement. Stops showed longer closure durations, longer voicing lags, and higher H1-H2 values in higher prosodic positions.

The role of segmental and intonational cues in dialect discrimination

Research indicates that adult listeners are able to use intonation to discriminate between two languages when one of the languages is familiar (Ramus and Mehler, 1999; Pijper, 1983). In this paper, we test adults to determine whether they use segmental or intonational cues to distinguish their native dialect from a foreign one. In three experiments, American English listeners were asked to categorize American and Australian English sentences when (a) segmental and supra‐segmental cues are available, (b) sentences are re‐synthesized with flat intonation, leaving only segmental cues, (c) segmental information is stripped away, leaving only intonation. Results will be discussed in the context of infant research demonstrating that five month olds are able to distinguish different dialects of the same language (Nazzi, Jusczyk, and Johnson 2000).

Role of rhythmic and intonational cues in language discrimination.

Languages differ in rhythm as well as intonation. Research indicates that adult listeners are able to use rhythm to discriminate between two languages from different rhythm classes [Ramus Mehler, (1999)]. For languages within the same rhythm class, adults are able to use to use intonation to discriminate between languages like English and Dutch, but only when one of the languages is familiar [Ramus and Mehler, (1999); Pijper, (1983)]. It remains unclear if the rhythmic differences between languages in the same rhythm class are enough to support language discrimination. In this paper, we tested American English listeners’ ability to categorize re‐synthesized American English and German sentences or American and Australian English sentences from which all segmental information had been removed. English and German are from the same rhythm class and differ in intonation; whereas American and Australian English can be thought to be rhythmically identical, but differ in intonation. Subjects were tested in three...

Weighting of prosodic cues in language discrimination by infants and adults

Previous research has shown that infants and adults can discriminate between prosodically similar languages using only prosodic cues. These experiments were designed to determine whether listeners use pitch cues or segmental duration and timing cues (i.e., rhythm cues) in language discrimination. We tested American English learning 7-month-old infants and adults on their ability to discriminate between sentences in American English and German that had been re-synthesized to isolate different cues. Infants were presented with low-pass filtered speech and speech with re-synthesized sinusoidal pitch. Adults were presented with low-pass filtered speech and two conditions where sonorants were replaced with /a/ and obstruents were replaced with silence: one contained both pitch and durational cues and the other contained only durational cues. Pitch cues were important for infants as well as adults. For infants, pitch cues were necessary for successful language discrimination; neither segmental nor durational cu...

Intonation in Tongan.

This paper presents a model of the intonational system of Tongan, an Austronesian language, taking the autosegmental 괂 metrical theory as its framework. Tongan has lexical stress which appears on the penultimate syllable of prosodic words and is marked post 괂lexically with one of two bitonal pitch accents—a rise, LH * , or a low tone, L * . Measurements show that the first tone of both pitch accents aligns with the stressed syllable onset, while the second tone aligns with the stressed syllable offset. There is evidence for two tonally marked levels of prosodic phrasing in Tongan, the intonational phrase (IP) and the accentual phrase (AP). The IP is about the size of a full utterance or major phrase and is marked by a final boundary tone and are realized on the IP 괂final syllable. Four boundary tones have been observed. The smaller unit, the AP, usually contains one lexical word plus preceding functional elements. Two AP 괂final tones have been observed, realized on the final syllable of the phrase. Lastly, focus is only realized intonationally through increased pitch range on the focused element. Tongan is typologically interesting because it provides another case in a growing list of languages that intonationally marks both head and edge prominence.

Language and dialect discrimination by five‐month‐olds.

To acquire a language, infants must be able to accurately discriminate speech utterances belonging to that language. This is particularly important for infants raised in bilingual environments, who must be able to tag utterances they hear as belonging to one language or another. From birth, infants are able to discriminate languages from different rhythm classes, but not languages within the same rhythm class [Nazzi et al. (1998)]. By 5 months, infants can discriminate two languages from the same rhythm class, as long as one language is native. Johnson [(2000)] showed that English‐learning infants could discriminate English from Dutch, and even American English from British English, but not Dutch from German. We attempt to extend these results to American English and German, and American and Australian English. Preliminary results show that English‐learning 5‐month‐olds cannot discriminate American English from German but are able to discriminate American from Australian English. These results fit with pr...

Examining the role of pitch, duration, and intensity in word segmentation.

English listeners can make use of stress cues in word segmentation. Stress is associated with a number of phonetic cues, including pitch movements, longer durations, and greater intensity. Studies on stress perception have shown that pitch is the most powerful cue to English stress, followed by increased duration and greater intensity [Fry (1958); Bolinger (1958)], suggesting that pitch alone might be sufficient to cue word boundary. Here, I test whether pitch alone is enough to cue word boundary for English listeners, using an artificial language paradigm. The artificial language used contains no distributional cues to word boundary, so words can only be segmented using pitch. I also pit pitch cues against intensity and durational cues in order to test whether English listeners weight cues for stress in a word segmentation task, or if they require correlates of stress to be bundled together.