The Phonetics Cluster is concerned with human spoken language, more specifically, the study of speech production and perception.
The study of speech production is concerned with the analysis of the acoustic signal as well as of articulatory movements, for example, of the tongue or lips. Among other things, it investigates how the sounds of a language or variety are generated, how they differ from other sounds, languages, or varieties, and how their productions vary between speakers. Another factor analyzed is voice quality to assess whether certain speakers or speaker groups have an especially creaky or breathy voice.
The study of speech perception is concerned with how listeners "pick out" individual sounds from the acoustic speech signal and combine them into words. A further subject of investigation is which acoustic properties are particularly important for listeners in recognizing certain sounds and, subsequently, words, and which properties allow listeners to recognize speakers as speakers of a certain variety, for example.
Our research aims to understand how human language works to test and expand theoretical models of speech production and perception, as well as linguistic theories. As part of our long-term strategic orientation, the Phonetics group contributes to the Lighthouse Project Adaptive Speech Perception.
Research Methods
The methods used in phonetics are manifold. Data are collected in the laboratory and in the field, with informants at work and at home. Depending on the object of study, speech production measurements are conducted using sound recordings (for acoustic analysis) or ultrasound tongue imaging (for articulatory analysis). Video recordings, face/lip tracking, and electroglottographic (EGG) measurements (e.g., for voice analysis) are also used. In studies on speech perception, participants are presented with natural or synthesized speech samples (e.g., sounds, syllables, words, or sentences). They are asked to indicate which sound or word they hear, how "well" a word is pronounced, and whether it is an actual word, etc. Both their answers and reaction times are then measured. Measurements of eye movements, what kind of speakers and/or images the listeners are looking at while listening to words or sentences, can provide information about how the speech signal is perceived in real time.
Cross-cluster Interfaces
The diversity of the Phonetics department leads to numerous interfaces and collaborations with other departments at the institute. For example, the modeling of articulation movements, acoustic features of sound categories, or the regional distribution of language varieties forms interfaces with mathematics and machine learning. The evaluation of speech intelligibility in noise combines questions of phonetics with the fields of numerical analysis and hearing. Questions of cognitive language processing also connect phonetics with the field of hearing. Comparing human linguistic communication with that of animals opens up opportunities for collaboration with biology.
Linguistic Variation and Listener Adaptation
Spoken language is highly variable. Sounds, words, and sentences can be produced in many different ways, yet speech perception usually feels effortless. The long-term goal of the Phonetics group is to understand how listeners adapt to this variation across various situations, enabling communication to function smoothly and appropriately.
Situations and Conversation Partners
By 'situations,' we refer to both communication with different conversation partners and environmental factors such as noise or distraction. Conversation partners may speak in their native language, a foreign language, or a familiar or unfamiliar dialect. Many people know the experience of initially struggling to understand an unfamiliar speaker, but improving noticeably after a short time. We investigate which kinds of information about speakers listeners make use of when recognizing spoken language. Such information may stem from the conversation's context, the structure of the language itself, or the acoustic characteristics of the sounds. Acoustic cues, such as speech rate and individual voice qualities, also play an important role.
Perception in Noise and Distraction
We also study how listeners adjust when exposed to background noise or when they are distracted by additional tasks. Our research examines the circumstances under which different parts of speech information are perceived more or less strongly. For example, listeners may rely more heavily on their knowledge of a speaker's individual characteristics, on their expectations about how words are structured in the language they hear, or on the acoustic signal itself.
Relevance and Applications
This research is relevant for understanding linguistic sound systems and their evolution, for modeling cognitive processes during social interaction, and for applied areas such as language learning and communication in people with cognitive or hearing impairments. It also includes examining how speakers themselves adapt during conversation, as their behavior is closely linked to how speech is perceived.
