PRICAI 2008: Trends in Artificial Intelligence 10th Pacific Rim International Conference on Artificial Intelligence, Hanoi, Vietnam, December 15-19, 2008. Proceedings

A speech recognition system for Chinese names based on Karhunen Loeve transform (KLT), MFCC, hidden Markov model (HMM) and Viterbi algorithm is proposed in this thesis. KLT is the optimal transform in minimum mean square error and maximal energy packing sense to reduce data. HMM is a stochastic approach which characterizes many of the variability in speech signal by recording the state transitions. For the speaker-dependent case, the correct identification rate can be achieved 93.97% within 3 seconds in the laboratory environment.

A Mandarin speech recognition system for addresses based on MFCC, hidden Markov model (HMM) and Viterbi algorithm is proposed in this thesis. HMM is a doubly stochastic process describing the ways of pronunciation by recording the state transitions according to the time-varing properties of the speech signal. In order to simplify the system design and reduce the computational cost, the mono-syllable structure information in Mandarin is used by incorporating both mono-syllable recognizor and HMM for our system. For the speaker-dependent case, Mandarin address inputting can be accomplished within 60 seconds and 98% correct identification rate can be achieved in the laboratory environment.

A novel phrase recognition method is proposed. It eliminates the speech difference between intraspeaker or interspeaker by transform phrases to difference subspace. A new endpoint detection method is also proposed, it can detection the human speech signal more effectively. All methods are test and verify at Microsoft Windows environment.

A Mandarin keyword spotting system based on MFCC, discrete-time HMM and Viterbi algorithm with DTW is proposed in this thesis. Joining with a dialogue system, this keyword spotting platform is further refined to a prototype of natural speech patient registration system of Kaohsiung Veterans General Hospital. After the ID number is asked by the computer-dialogue attendant in the registration process, the user can finish all relevant works in one sentence. Functions of searching clinical doctors, making and canceling registration are all built in this system. In a laboratory environment, the correct rate of this speaker-independent patient registration system can reach 97% and all registration process can be completed within 75 seconds.

The objective of this thesis is to build a phrase recognition system under noisy environment that can be used in real-life. In this system, the noisy speech is first filtered by the enhanced spectral subtraction method to reduce the noise level. Then the MFCC with cepstral mean subtraction is applied to extract the speech features. Finally, hidden Markov model (HMM) is used in the last stage to build the probabilistic model for each phrase. A Mandarin microphone database of 514 company names that are in Taiwan¡¦s stock market is collected. A speaker independent noisy phrase recognition system is then implemented. This system has been tested under various noise environments and different noise strengths.

This dissertation describes six syntactic complex predicates in Q'anjob'al (Maya) spoken in Santa Eulalia, Huehuetenango, Guatemala: the Directional Construction (DIRC), Verbal Resultative Predicate (resultative V1V2), Causative Predicate (causative V1V2), Complement-like Predicate (complement-like V1V2), Preverbal Resultative Predicate (R2°P), and Positional Resultative Construction (PRC). They resemble resultatives, serial verbs, and causatives in other languages. The dissertation describes their monoclausal structure, meaning, event and argument structure, and lexical restrictions. These translate into five parameters: (a) they have a single value of time, aspect, modality and polarity, (b) they have a single set of arguments, (c) they have one intonational contour, (d) the elements denote a single event or subparts of a macro event, and (e) the elements minimally involve argument sharing, but could also involve argument fusion, or composition. Q'anjob'al complex predicates have two main features: (a) they are of an asymmetric type in that one element functions as the 'primary' element, and (b) they always have a verbal element (V) with another element that can be verbal or nonverbal (NV). Thus, these complex predicates always have a verbal 'primary' element. Regarding their structure, complex predicates involve four constructions: (a) a nonverbal-verbal construction used by the R2°P and shared by depictives (where it is a non-complex predicate and multi-headed clause), (b) a verb-verb construction used for causative, resultative, and complement-like V1V2s and shared by DIRCs, (c) a verbdirectional construction, a serial verb type, where V is the main verb and DIR corresponds to up to three motion verbs, and (d) a verb-positional construction used by the PRC. In their argument structure, resultatives and causative V1V2s, PRCs, and R2°Ps involve argument fusion, complement-like V1V2s involve raising, and DIRCs may involve argument sharing or fusion depending on the particular type. Regarding event structure, resultative V1V2s, and aspectual DIRCs denote a single event, and other complex predicates denote macro events. Finally, lexical semantics is central for distinguishing complex predicates from each other and from other multi-headed clauses. This dissertation contributes to the documentation of Q'anjob'al and advances the syntactic and semantic analysis of Mayan languages. It also contributes to our understanding of complex predicates through a case study. / text

Argument marking prepositions in German are part of more complex structures referred to here as verb-preposition combinations (verb-PPs), e.g. warten auf ('to wait for') and pochen auf ('to insist on'). The preposition auf ('on') attaches to a wide range of verbs to form such combinations in which auf encodes different semantic relations that elude concrete description. Nevertheless, previous research in valency theory and related approaches could identify patterns in the distribution of verb-PP[subscript 'auf'] combinations (Eroms 1981, 1991, Lerot 1982, Bouillon 1984, Domínguez Vázquez 2005), based on perceived similarities in the meaning of the governing verbs. Cognitive linguistics provides insights into seemingly opaque senses of prepositions by analyzing them as motivated by metaphorical meaning extension (Brugman 1988, Lakoff 1987, Meex 2001, Liamkina 2007). Finally, generative approaches scrutinize the semantic relationships between verbs and their PP-arguments and systematize them under the concept of semantic roles (Fillmore 1968, Rauh 1993). However, none of these approaches can fully account for the distribution of verb-PPauf combinations in German. This dissertation proposes a novel approach towards identifying and analyzing the distributional patterns of verb-PP[subscript 'auf'] combinations by applying insights from Frame Semantics (Fillmore 1982, 1985) and Construction Grammar (Goldberg 1995, 2006). Goldberg's theory of argument structure constructions already served as a model for analyzing auf as a partially schematic argument structure construction encoding the meaning 'future orientation/future event' (Rostila 2007). Based on a large amount of corpus data, I show that such generalizing accounts are better arrived at by employing a usage-based bottom-up approach to verb-PP[subscript 'auf'] combinations. I argue that the detailed semantic and syntactic information provided by the lexical database FrameNet for each lexical unit can be used to identify distributional patterns and to describe them in detail. Furthermore, I argue that integrating the verb-PP[subscript 'auf'] combinations and the frames they evoke into a hierarchical lexical-constructional network allows us to discover substantiated generalizations about these combinations while at the same time preserving the description of their idiosyncratic features. / text

published_or_final_version / abstract / toc / Linguistics / Master / Master of Philosophy

The first part of this thesis addresses the following questions: where in the syntactic tree, and at what representational level is an NP Case-checked. To this end, it presents converging data from French, Welsh and Irish, which suggest (i) that Case-checking may be accomplished under a variety of functional projections (subject to parametric variation); and (ii) that Case positions are--at least partially--independent of the A/A$ sp prime$-distinction. It furthermore presents evidence from Irish and Welsh--VSO languages in which NPs typically raise to their Case position only at LF--that NPs are, under certain conditions, Case-checked at S-structure. / Chapter 2 investigates word order and cliticisation in Standard French and Quebec French interrogatives and proposes a typology of interrogatives. Chapter 3 and 4 account for complementizer variation, pre-verbal particles and agreement patterns in Welsh and Irish under a Case-theoretic approach. / The second part of this thesis concerns the conditions on the availability of structural accusative Case. A theory of structural Case is proposed according to which accusativity is a configurational rather than a lexical property--i.e., resulting from syntactic geometry and not from lexical feature specifications on verbs. To this end, a comparison between the syntactic mapping of stative and perfective predicates in Irish and English is undertaken.