We are interested in on-line acquisition of acoustic, lexical and semantic units from spontaneous speech. Traditional ASR techniques require the domain-speci c knowledge of acoustic, lexicon data and more importantly the word probability distributions. In this paper we propose an algorithm for unsupervised learning of acoustic and lexical units from out-of-domain speech data. The new lexical un...

The semantic annotation of corpora has an important role to play in ensuring that sentences occurring in natural language texts are correctly understood based on their intended context. Two examples of lexical semantic units that contribute to this knowledge are word senses – which allow words with multiple meanings to be understood based on the context in which they are used – and named entiti...

It is generally assumed that during reading, the activation produced over orthographic units feeds forward to phonological units. Supporting interactive models of word recognition, Stone, Vanhoy, and Van Orden (1997) recently claimed that phonological activation reverberates to orthographic processing units and consequently constrains orthographic encoding. They found that the consistency of th...

Lexical unit is a word or collocation. Extracting lexical knowledge is an essential and difficult task in NLP. The methods of extracting of lexical units are discussed. We present a method for the identification of lexical boundaries. The problem of necessity of large corpora for training is discussed. The advantage of identification of lexical boundaries within a text over traditional window m...

One of the key challenges involved in building statistical automatic speech recognition (ASR) systems is modeling the relationship between subword units or “lexical units” and acoustic feature observations. To model this relationship two types of resources are needed, namely, acoustic resources i.e., speech data with word level transcriptions and lexical resources where each word is transcribed...

How do we map the rapid input of spoken language onto phonological and lexical representations over time? Attempts at psychologically-tractable computational models of spoken word recognition tend either to ignore time or to transform the temporal input into a spatial representation. TRACE, a connectionist model with broad and deep coverage of speech perception and spoken word recognition pheno...

The present corpus-based lexical study reports the development of a Pharmacy Academic Word List (PAWL); a list of the most frequent words from a corpus of 3,458,445 tokens made up of 800 most recent pharmacy texts including research articles, review articles, and short communications in four sub-disciplines of pharmacy. WordSmith (Scott, 2017) and AntWordProfiler (Anthony, 2014) were used to sc...

German is a highly inflected language with a large number of words derived from the same root. It makes use of a high degree of word compounding leading to high Out-of-vocabulary (OOV) rates, and Language Model (LM) perplexities. For such languages the use of sub-lexical units for Large Vocabulary Continuous Speech Recognition (LVCSR) becomes a natural choice. In this paper, we investigate the ...

One of the largest remaining unsolved mysteries in cognitive science is how the rapid input of spoken language is mapped onto phonological and lexical representations over time. Attempts at psychologically-tractable computational models of spoken word recognition tend either to ignore time or to transform the temporal input into a spatial representation. This is the approach taken in TRACE (McC...