Computational semantics performs automatic meaning analysis of natural language. Research in computational semantics designs meaning representations and develops mechanisms for automatically assigning those representations and reasoning over them. Computational semantics is not a single monolithic task but consists of many subtasks, including word sense disambiguation, multi-word expression analysis, semantic role labeling, the construction of sentence semantic structure, coreference resolution, and the automatic induction of semantic information from data. The development of manually constructed resources has been vastly important in driving the field forward. Examples include WordNet, PropBank, FrameNet, VerbNet, and TimeBank. These resources specify the linguistic structures to be targeted in automatic analysis, and they provide high-quality human-generated data that can be used to train machine learning systems. Supervised machine learning based on manually constructed resources is a widely used technique. A second core strand has been the induction of lexical knowledge from text data. For example, words can be represented through the contexts in which they appear (called distributional vectors or embeddings), such that semantically similar words have similar representations. Or semantic relations between words can be inferred from patterns of words that link them. Wide-coverage semantic analysis always needs more data, both lexical knowledge and world knowledge, and automatic induction at least alleviates the problem. Compositionality is a third core theme: the systematic construction of structural meaning representations of larger expressions from the meaning representations of their parts. The representations typically use logics of varying expressivity, which makes them well suited to performing automatic inferences with theorem provers. Manual specification and automatic acquisition of knowledge are closely intertwined. Manually created resources are automatically extended or merged. The automatic induction of semantic information is guided and constrained by manually specified information, which is much more reliable. And for restricted domains, the construction of logical representations is learned from data. It is at the intersection of manual specification and machine learning that some of the current larger questions of computational semantics are located. For instance, should we build general-purpose semantic representations, or is lexical knowledge simply too domain-specific, and would we be better off learning task-specific representations every time? When performing inference, is it more beneficial to have the solid ground of a human-generated ontology, or is it better to reason directly with text snippets for more fine-grained and gradual inference? Do we obtain a better and deeper semantic analysis as we use better and deeper manually specified linguistic knowledge, or is the future in powerful learning paradigms that learn to carry out an entire task from natural language input and output alone, without pre-specified linguistic knowledge?
Acoustic Theories of Speech Perception
Melissa Redford and Melissa Baese-Berk
Acoustic theories assume that speech perception begins with an acoustic signal transformed by auditory processing. In classical acoustic theory, this assumption entails perceptual primitives that are akin to those identified in the spectral analyses of speech. The research objective is to link these primitives with phonological units of traditional descriptive linguistics via sound categories and then to understand how these units/categories are bound together in time to recognize words. Achieving this objective is challenging because the signal is replete with variation, making the mapping of signal to sound category nontrivial. Research that grapples with the mapping problem has led to many basic findings about speech perception, including the importance of cue redundancy to category identification and of differential cue weighting to category formation. Research that grapples with the related problem of binding categories into words for speech processing motivates current neuropsychological work on speech perception. The central focus on the mapping problem in classical theory has also led to an alternative type of acoustic theory, namely, exemplar-based theory. According to this type of acoustic theory, variability is critical for processing talker-specific information during speech processing. The problems associated with mapping acoustic cues to sound categories is not addressed because exemplar-based theories assume that perceptual traces of whole words are perceptual primitives. Smaller units of speech sound representation, as well as the phonology as a whole, are emergent from the word-based representations. Yet, like classical acoustic theories, exemplar-based theories assume that production is mediated by a phonology that has no inherent motor information. The presumed disconnect between acoustic and motor information during perceptual processing distinguishes acoustic theories as a class from other theories of speech perception.