SimPhon.Net - Usage-based approaches, computational modeling and simulation studies in phonetics and phonology

点击翻译按钮获取中文摘要

Two kinds of research methodology are often discussed in phonetic sciences: field work, observing its subject ofstudy in situ, and experimental or laboratory work, studying speech in vitro. Both observe language in vivo onliving, speaking humans. This network is concerned with a third kind of research method that recently emerged:the study of language and speech by means of computational modeling and simulations, observing the subjectof study in silico. Fant [2004] noticed a transition from analogue to digital processing in speech research. Thistrend towards digitization is still in progress, manifesting itself in the emergence of simulation-based methodology.However, simulations are not considered common practice, yet. They require familiarity with artificial intelligenceand machine learning. Thus, such experiments can be performed only by a phonetician or phonologist with astrong background in computational linguistics, or in close interdisciplinary collaboration between linguists andcomputer scientists. SimPhon.Net will address that challenge, each member contributing their expertise and experience. Combining this knowledge within the network, we will review and advance on the state of the art, and identify most promising approaches and most important challenges.