Für diesen Text ist leider keine deutsche Übersetzung verfügbar.
This area deals with acoustic, physiological, psychoacoustic and computational aspects of phonetics and linguistics.
The acoustic source of voiced speech sounds is located at the vocal folds (glottis). During phonation a quite complex interaction between the control of the subglottal air pressure and the muscles of the glottis takes place. The periodic opening and closing of the vocal folds produces air pressure pulses generating a harmonic source spectrum with decaying amplitudes of the partials from the fundamental frequency by -12dB/octave. The acoustic radiating characteristic at the lip opening provides a gain of +6 dB/octave that the net decay of the amplitude spectrum results at -6dB.
The pharynx and and the oral cavity act as adjustable resonance volumes shaping the glottal source spectrum to the known vowel sounds (timbre). In case of the lowered velum the nasal cavity is coupled to the vocal tract influencing the spectrum of voiced sounds.
Individual and social life finds its expression in speech. Even subtle, prosodic and segment differences in speech are perceived. The monophthongisation of dipthongs (/aE/ und /aO/) is the result of a century's progressive change. The development of the monophthongisation is both individually and regionally specific. The first fomant graph (F1, F2) shows two Viennese speakers, black dipthongs (D) and red monopthongs (M). The second format graph shows 2 Lower Styrian speakers (black dipthongs (D) and red monopthongs (M)). The /O/ quality (Vienna, M) and the /A/ quality (Lower Styria, M) are audible.
The impact of emotion on speech behavior is investigated in both studio and real-life recordings.
During a long-standing cooperation with the Commission for linguistics and communication sciences of the Austrian Academy of Sciences, a huge amount of material on Austrian German has been collected.
Phonological and phonetic reduction processes in Austrian German and in several other languages, including Albanian, Serbian, Romanian, Italian and Tibetan, is being investigated; formant measurements for vowel quality considerations are measured.
The Acoustics Research department has acquired over the years a huge amount of real-life recordings of Austrian dialect speakers (more than 200 different speakers); this material is used for sociophonetic as well as emotion and speech studies.
Several students are currently working on foreigner talk, especially with L1: Polish / L2: German.
First language acquisition
A relational child language database is being built containing full transcriptions in orthographic and phonetic format and several other devices for corpus analyses such as phonological and morphological information. Apart from the entries of the database are linked with the corresponding digital sound signals so that immediate access to the original data as well as automatic phonetic analyses are possible. The database will mainly contain data from the child Bernd (Austrian German) but also from other children speaking French, Polish and various other languages.
Formant Candidate Extraction LPC-spectrogram with peak detection (to probe further: see S_TOOLS-STx)