• 대한음성학회지:말소리
• /
• 제53호
• /
• pp.75-92
• /
• 2005

Prosody Generation Based on C-ToBI Representation for Text-to-SpeechSeungwon Kim, Yu Zheng, Gary Geunbae Lee, Byeongchang KimProsody modeling is critical in developing text-to-speech (TTS) systems where speech synthesis is used to automatically generate natural speech. In this paper, we present a prosody generation architecture based on Chinese Tone and Break Index (C-ToBI) representation. ToBI is a multi-tier representation system based on linguistic knowledge to transcribe events in an utterance. The TTS system which adopts ToBI as an intermediate representation is known to exhibit higher flexibility, modularity and domain/task portability compared with the direct prosody generation TTS systems. However, the cost of corpus preparation is very expensive for practical-level performance because the ToBI labeled corpus has been manually constructed by many prosody experts and normally requires a large amount of data for accurate statistical prosody modeling. This paper proposes a new method which transcribes the C-ToBI labels automatically in Chinese speech. We model Chinese prosody generation as a classification problem and apply conditional Maximum Entropy (ME) classification to this problem. We empirically verify the usefulness of various natural language and phonology features to make well-integrated features for ME framework.

• 말소리와 음성과학
• /
• 제15권1호
• /
• pp.47-54
• /
• 2023

This study attempted to test the hypothesis that Korean evaluators can score L2 speech appropriately, even when speech rate features are unavailable. Two perception experiments-preliminary and main-were conducted sequentially. The purpose of the preliminary experiment was to categorize English-as-a-foreign-language (EFL) speakers into two groups-advanced learners and lower-level learners-based on the proficiency scores given by five human raters. In the main experiment, a set of stimuli was prepared such that the speech rate of all data tokens was modified to have a uniform speech rate. Ten human evaluators were asked to score the stimulus tokens on a 5-point scale. These scores were statistically analyzed to determine whether there was a significant difference in utterance production between the two groups. The results of the preliminary experiment confirm that higher-proficiency learners speak faster than lower-proficiency learners. The results of the main experiment indicate that under controlled speech-rate conditions, human raters can appropriately assess learner proficiency, probably thanks to the linguistic features that the raters considered during the evaluation process.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권1호
• /
• pp.297-306
• /
• 2024

At a specific juncture in industrial development, animation emerges as a product imbued with discernible features characteristic of the mechanical replication era. Through the integration of technology, it continually gives rise to postmodern cultural traits. Notably, the trajectories of postmodernism in China and Western Europe exhibit divergence. Following the reform and opening up, both postmodernism and modernism found acceptance and became interwoven. Chinese animated films, influenced by postmodernism, actively challenge traditional animation norms, exemplifying postmodern characteristics such as deconstruction, intertextuality, indeterminacy, collage, and diversity. This article endeavors to delve into the discourse surrounding postmodernism in both Western and Chinese contexts. It aims to scrutinize the manifestation of postmodern features, particularly with regard to ethnic elements, traditional mythological texts, and gender representations in Chinese animated films. Consequently, it becomes imperative to contemplate the affirmative adaptation and independent development of postmodernism within the unique Chinese milieu. Building upon an analysis of noteworthy traditional animated film genres and developmental trends in China, this study focuses on animated films from the 21st century, specifically Monkey King: Hero is Back, Big Fish & Begonia, Dahufa, White Snake and Nezha: Birth of the Demon Child. The paper seeks to explore the overarching characteristics of postmodernity reflected in the linguistic environment of postmodernism in China.

• 영어어문교육
• /
• 제17권3호
• /
• pp.255-272
• /
• 2011

As the demand of English education is increasing, the demand for Native English speaking teachers (NEST) is rising, especially in Asian countries. However, due to the low number of NEST, the Korean government is suggesting that Filipino Women be used as English teachers as an alternative. This study aims to answer three questions: (1) are Filipino women in Korea qualified to teach English based on the error analysis of their written essays? (2) what are the linguistic features found in their diagnostic essays? and (3) is their written English better than Korean college students' written English based on the comparison of the two groups? The findings from the Filipino participants show the most frequently occurring errors are related to punctuation usage (commas and hyphens), vocabulary (word choice), verb usage, redundancy, and even as basic as capitalization usage. The results from the comparison of the two groups show that the percentage of the Filipino participants' written error was 14% while the percentage of the Korean participants was 17%. The findings would give us some ideas on the qualification of Filipino women in Korea as English teachers.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1999년도 학술발표대회 논문집 제18권 2호
• /
• pp.337-340
• /
• 1999

The most salient and discussed features of speech production in agrammatic aphasia are the omission and substitution of grammatical morphemes. Cross-linguistic studies have shown that the pattern of omission/substitution is not random but occurs in a systematic and highly constrained way. Although these descriptions are important, they do not explain why all grammatical morphemes are not equally impaired. Friedmann and Grodzinsky (1997) proposed the Tree-Pruning Hypothesis (TPH) to account for these patterns of sparing and loss. The TPH claims that in an agrammatic representation, an impaired functional node is underspecified, thus allowing inappropriate affixation to occur. Additionally, whenever a node is impaired, all nodes above it will also be impaired. Using four types of narratives collected from two Korean agrammatic patients, We test the claim that the impairment in agrammatism is based on such hierarchical representation. It was found that these patients consistently produced appropriate grammatical morphemes that are higher in a syntactic tree than the impaired morphemes. The finding that an intact node exists higher than an impaired node refutes the TPH.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.203-207
• /
• 1998

In this paper, it is shown how Fuzzy Pattern Matching can be applied to diagnosis of the most common faults of Rotating Machinery. The whole diagnosis process has been divided in three steps : Fault Detection, Fault Isolation and Fault Identification, whose possible results are described by linguistic patterns. Diagnosis will consist in obtaining a set of matching indexes that indexes that express the compatibility of the fuzzified features extracted from the measured vibration signals, with the knowledge contained in the corresponding patterns.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2002년도 Language, Information, and Computation Proceedings of The 16th Pacific Asia Conference
• /
• pp.157-165
• /
• 2002

This paper describes our ongoing Korean-Chinese machine translation system, which is based on verb patterns. A verb pattern consists of a source language pattern part for analysis and a target language pattern part for generation. Knowledge description on lexical level makes it easy to achieve accurate analyses and natural, correct generation. These features are very important and effective in machine translation between languages with quite different linguistic structures including Korean and Chinese. We performed a preliminary evaluation of our current system and reported the result in the paper.

• 한국어정보학회:학술대회논문집
• /
• 한국어정보학회 2016년도 제28회 한글및한국어정보처리학술대회
• /
• pp.85-89
• /
• 2016

프레임넷 (FrameNet) 프로젝트는 버클리에서 1997년에 처음 제안했으며, 최근에는 다양한 언어적 특징을 반영하여 여러 국가에서 사용되고 있다. 하지만 문장의 프레임을 분석하는 것은 자연언어처리 전문가들이 많은 시간을 들여야 한다. 이 때문에, 한국어 프레임넷을 처음 만들 때는 충분한 훈련을 받은 번역가들이 영어 프레임넷의 문장들과 그 주석 정보들을 직접 번역하는 방법을 사용했다. 결과적으로 상대적으로 적은 비용이 들지만, 여전히 한 문장에 여러 번 등장하는 프레임 정보를 모두 번역하고 에러를 분석해야 했기에 많은 노력이 들어갔다. 본 연구에서는 일본어와 한국어의 언어적 유사성을 사용하여 비교적 적은 비용으로 한국어 프레임넷을 확장하는 방법을 제시한다. 또한 프레임넷에 친숙하지 않은 사용자가 더욱 쉽게 프레임 정보를 활용할 수 있도록 PubAnnotation 기술을 도입하고 "조사"라는 특성을 고려한 Valence pattern 분류를 통해 한국어 공개 프레임넷 사이트를 개선하였다.

• 대한음성학회지:말소리
• /
• 제66호
• /
• pp.41-59
• /
• 2008

Knowledge in linguistic rhythm of the target language plays a major role in foreign language proficiency. This study attempts to discover valid rhythm features that can be utilized in automatic assessment of non-native English pronunciation. Eight previously proposed and two novel rhythm metrics are investigated with 360 English read speech tokens obtained from 27 Korean learners and 9 native speakers. It is found that some of the speech-rate normalized interval measures and above-word level metrics are effective enough to be further applied for automatic scoring as they are significantly correlated with speakers' proficiency levels. It is also shown that metrics need to be dynamically selected depending upon the structure of target sentences. Results from a preliminary auto-scoring experiment through a Multi Regression analysis suggest that appropriate control of unexpected input utterances is also desirable for better performance.

• 한국영어학회지:영어학
• /
• 제4권1호
• /
• pp.83-102
• /
• 2004

The purpose of this paper is to find out some major characteristics of intermediate-advanced Korean learners' English by corpus- linguistically analyzing their essays in comparison with native speakers'. We construct a corpus of CBT TOEFL essays by Korean learners, NNS1 (94076 words in 402 texts), and its sub-corpus, NNS2 (14291 words in 45 texts), and then a corpus of model essays written or meticulously edited by native speakers, NS (14833 words in 35 texts). We compare NNS1 and NNS2 with NS, and with some other corpora, in terms of high-frequency words, and show that Korean learners' writings have more features of informal writing than those of formal writing, which is in accord with the reports in Granger (1998) that EFL writings by European advanced learners are characterized by informality.