Acoustic phonetics involves the study of speech production and perception in regard to higher grammatical levels, especially phonology. The acoustic model of speech production computes the resonances of the vocal tract during different articulatory configurations. Formant frequencies, fundamental frequency, amplitude spectra and duration are used to describe language varieties phonetically. The results are interpreted within a phonological framework. The focus of our group lies on insufficiently described languages and language varieties.
We also develop new speech synthesis technology where our phonetic analyses are applied. The technologies range from acoustic and audio-visual speech synthesis, dialect and accent transformation to singing synthesis. In dialect synthesis we developed methods to generate in-between varieties that lie between a standard and a dialect, and thereby allow for a more realistic speech output production. Visual speech synthesis allows for the animation of a virtual character from textual input.
Further Information
- Speech and Language Processing
- Acoustic Model of Speech Production (in German)
- Speech synthesis of Viennese dialects
- Speech data resources of the Acoustic Phonetics group
Running projects
- Variation and Change of dialect varieties in Austria (in apparent and real time)
- Vowel and consonant quantity in Southern German varieties
- Lateral Variants of Bosnian Migrants Living in Vienna
- Speech Synthesis of Dialects and Language Varieties
- Synthesis of Singing
- Acoustic cues in production and perception of irony in normal-hearing and CI-listeners
Completed projects
- Gestural coordination and speech rhythm
- Phonetics and Phonology of the Viennese Dialect
- Die österreichische Standardaussprache Wiens in Kontakt mit der deutschen Standardaussprache
- Vowel Systems in Comparison
- Acoustic Features for Speaker Models
- Automatic Speaker Identification
- Automatic Speaker Verification
- Soziolekte in Wien - die mittelbairischen Varietäten
- Adaptive Audio-Visual Dialect Speech Synthesis
- Machine Learning Techniques for Modelling Language Varieties
- Optimal Gaussian Mixture Model (GMM) Initialization for Speaker Modeling
- Speaker Clustering
- The Vowels of Standard Austrian German
- Vowel Tensity in Standard Austrian and Standard German
- BE-SyMPHONic
Staff
PD Dr. Eva Reinisch, Head of the Group