• Language and Information
• /
• 제14권2호
• /
• pp.103-129
• /
• 2010

In this paper, I show that anaphora across speakers shows both dynamic and static sides. To capture them all formally, I will adopt semantics based on the assumption that variables range over individual concepts that connect epistemic alternatives. As information increases, a variable can take a different range of possible individual concepts. This is captured by the notion of virtual individual (= vi), a set of individual concepts which are indistinguishable in an information state. The use of a pronoun involves two information states, one for the antecedent, which is always part of the common ground, and the other for the pronoun. Information increase changes vis for variables in the common ground. A pronoun can be used felicitously if there is a unique virtual individual in the information state for the antecedent which does not split in two or more distinctive virtual individuals in the information state for the pronoun. The felicity condition for cross-speaker anaphora can be satisfied in declaratives involving modality, interrogatives and imperatives in a rather less demanding way, because in these cases the utterance does not necessarily require non-trivial personal information for proper use of a pronoun.

• The Journal of the Acoustical Society of Korea
• /
• 제16권3E호
• /
• pp.19-23
• /
• 1997

The goal of postprocessing in automatic speech recognition is to improve recognition performance by utterance verification at the output of recognition stage. It is focused on the effective rejection of out-of vocabulary words based on the confidence score of hypothesized candidate word. We present two methods for computing confidence scores. Both methods are based on the distance between each observation vector and the representative code vector, which is defined by the most likely code vector at each state. While the first method employs simple time normalization, the second one uses a normalization technique based on the concept of on-line garbage mode[1]. According to the speaker independent isolated words recognition experiment with discrete density HMM, the second method outperforms both the first one and conventional likelihood ratio scoring method[2].

• Language and Information
• /
• 제19권1호
• /
• pp.103-117
• /
• 2015

Not fully satisfied with the treatment of the so-called two nows in Korean by Lee & Choi (2009), this article seeks to furnish the issue with a firmer ground to base on in the relevant conversation. A close comparison between now and cikum appearing in the present perfect and present tense results in the two findings that (i) a crucial difference between the two adverbs is that Korean cikum lacks English now's ability to be identified with the reference time and (ii) further, seeming differences between them are not real but in fact due to tense and aspectual discrepancies between English and Korean. Thus, it claims, contra Lee (1976) and Park (2004), that cikum is a temporal locating adverb which invariably locates the event time of a given eventuality at the utterance time. In particular, it motivates that a past-tensed sentence with cikum should be understood as holding in the recent past mainly from pragmatic inferences rather than semantic entailments.

• Language and Information
• /
• 제17권1호
• /
• pp.17-32
• /
• 2013

Birner (1998) analyzes the construction of inversion within the centering theory, claiming that the preposed constituent in the inversion structure represents the backward-looking center that connects the current utterance to the previous discourse. However, this paper refutes such a strong claim, pointing out various problems of her work. Instead, this paper argues that the preposed element in the inversion construction is merely the preferred center under the condition that the ranking of the forward-looking centers is determined by the surface word order, rather than by grammatical relations. Thus, this paper claims that the discourse function of the construction of inversion is not text development but merely prominence-giving, in the sense of Ili$\acute{c}$ (1998).

• Proceedings of the KSPS conference
• /
• 대한음성학회 2006년도 춘계 학술대회 발표논문집
• /
• pp.33-36
• /
• 2006

This paper describes a speech recognizer implemented on PDAs. The recognizer consists of feature extraction module, search module and utterance verification module. It can recognize 37 words that can be used in the telematics application and fixed-point operation is performed for real-time processing. Simulation results show that recognition accuracy is 94.5% for the in-vocabulary words and 56.8% for the out-of-task words.

• Proceedings of the KSPS conference
• /
• 대한음성학회 1996년도 10월 학술대회지
• /
• pp.224-228
• /
• 1996

This paper is intended as a study on how an utterance is divided into rhythmic units in Standard Korean with respect to its syntactic structure. With respect to the data in this study I used 150 sentences which contained similar number of words and various syntactic structures. Those sentences were read by 7 speakers of Seoul dialect in a conversation style. Each sentence was read twice in a normal speed and twice in a fast speed. As a total, 4200 sentences were recorded. Then listening to them, the author marked the sentences with two kinds of boundaries i.e. strong and weak. To explore the relationship between rhythmic units and syntactic structure I devised a framework of grammatical symbols. Each symbol is designed to have both syntactic and morphological information at the same time. So I assigned those grammatical symbols to the sentences. Having sentences marked with grammatical symbols on the one hand, and with the rhythmic boundaries on the other hand, 1 could show the relationship between rhythmic units and syntactic structure; which syntactic structures are likely to be pronounced as one rhythmic unit, and which are on the rhythmic boundaries.

• MALSORI
• /
• 제40호
• /
• pp.13-27
• /
• 2000

It is generally accepted that there appears the rising tone at the end of the intermediate phrase in standard Korean. There have been discussions about whether the syllable with the rising tone, even if it is a particle or an ending, might be accented or not. The accented syllable is the most prominent one in the given phonological strings. It is determined by the nondistinctive stress which is located on the first or second syllable of lexical word according to vowel length and syllable weight. So pitch does not have any close relationship with accent. The intermediate phrase-final rising tone, therefore, is not associated with accent, but used to convey other pragmatic meanings, that is, i) speech style is more friendly, ii) the speaker tries to send the information for the hearer to hear more clearly, and iii) the speaker wants the hearer to keep on listening to him or her because the speaker's utterance is not complete.

• Phonetics and Speech Sciences
• /
• 제9권3호
• /
• pp.41-47
• /
• 2017

The purpose of this study is to examine how speakers use VOT and f0 to distinguish tense, lax, and aspirated stops in isolated sentence reading and paragraph readings. To do so, a total of 20 males between the ages of 20-25 years old were asked to read (1) isolated sentences, (2) information-oriented text and (3) emotional expressive texts in which the stop pronunciation's VOT value and f0 were measured thereafter. The main results are as follows. In the isolate sentence reading, lax stops, and aspirated stops were distinguished by both VOT and f0, but for the Korean men that read reading texts, VOT is not a cue to distinguish between lax and aspirated stops. In general, the VOT differences between lax stops and aspirated stops were smaller for information-oriented texts and emotional expressive texts than that of the isolate sentence reading. In the paragraph reading that induces a natural utterance, the f0 dependence is greater for the distinction between lax and aspirated stops.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제8권1호
• /
• pp.213-219
• /
• 2004

This paper proposes speaker recognition algorithm which includes the pitch parameter for the peak and valley. The time-frequency hybrid method for pitch extraction is valuable in that it can improve resolution in the time domain and accuracy in the frequency domain at the same time. It makes reference pattern using membership function and performs vocal track recognition of common character using fuzzy pattern matching in order to include time variation width for non-linear utterance for proposed method, speaker recognition experiments are carried out using vowels and number sounds.

• Speech Sciences
• /
• 제15권4호
• /
• pp.85-96
• /
• 2008

Automatic speech recognition has severe performance degradation under noisy environments. To cope with the noise problem, many methods have been proposed. Most of them focused on noise-robust features or model adaptation. However, researchers have overlooked utterance verification (UV) under noisy environments. In this paper we discuss UV problems based on the normalized confidence measure. First, we show that UV performance is also degraded in noisy environments with the experiments of an isolated word recognition. Then we observe how the degradation of UV performances is suffered. Based on the UV experiments we propose a modeling method of the statistics of phone confidences using sigmoid functions. For obtaining the parameters of the sigmoidal models, the particle swarm optimization (PSO) is adopted. The proposed method improves 20% rejection performance. Our experimental results show that the PSO-NCM can apply noise speech recognition successfully.