• Cross-Cultural Studies
• /
• v.46
• /
• pp.85-112
• /
• 2017

Donald Trump, the $45^{th}$ president of the United States, has revived the 'English Only' policy since the beginning of his presidential campaign. The monolingualism not only underscores his extremely conservative ideas, but it also reflects the nativist tendency that prevents the demographic and cultural transformation of the US, which is accelerated by globalization and transnational migration. In particular, Donald Trump tries to reconfirm the mainstream American culture that is now thought to have been threatened by Hispanization and the growing number of Spanish speakers. This paper examines the effects of "code-switching" and the possibility of a bilingual community by contrasting Donald Trump with Gloria $Anzald{\acute{u}}a$, one of the representative Latina writers who created a "border language." Borderlands/La Frontera (1987) includes Spanish glossaries and expressions to represent her bilingual realities, while attempting to translate from English to Spanish, and vice versa. However, the text occasionally demonstrates the impossibility of translation. In doing so, $Anzald{\acute{u}}a$ indirectly states that it is indispensable to present both languages at the stage; she also invites monolingual readers to make more efforts to learn and better understand the Other's language. A "border language" she attempts to embody throughout the text is created in the process of encounters, conflicts, and negotiations among languages of different ethnicities, classes and generations. It does not signify an established form: rather it appears as a constantly transforming language, which can provide us with new perspectives and an alternative way of communication beyond monolingualism.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.313-323
• /
• 2021

The purpose of this study is to propose rules that can identify the speaker referred by the speaker directive in the text for the realization of a machine learning-based virtual character using a literary text. Through previous studies, we found that when applying literary texts to machine learning, the machine did not properly discriminate the speaker without any specific rules for the analysis of speaker directives such as other names, nicknames, pronouns, and so on. As a way to solve this problem, this study proposes 'nine rules for finding a speaker indicated by speaker directives (including pronouns)': location, distance, pronouns, preparatory subject/preparatory object, quotations, number of speakers, non-characters directives, word compound form, dispersion of speaker names. In order to utilize characters within a literary text as virtual ones, the learning text must be presented in a machine-comprehensible way. We expect that the rules suggested in this study will reduce trial and error that may occur when using literary texts for machine learning, and enable smooth learning to produce qualitatively excellent learning results.

• Phonetics and Speech Sciences
• /
• v.16 no.1
• /
• pp.67-76
• /
• 2024

Recent advances in text-to-speech (TTS) technology have significantly improved the quality of synthesized speech, reaching a level where it can closely imitate natural human speech. Especially, TTS models offering various voice characteristics and personalized speech, are widely utilized in fields such as artificial intelligence (AI) tutors, advertising, and video dubbing. Accordingly, in this paper, we propose a one-shot multi-speaker TTS system that can ensure acoustic diversity and synthesize personalized voice by generating speech using unseen target speakers' utterances. The proposed model integrates a speaker encoder into a TTS model consisting of the FastSpeech2 acoustic model and the HiFi-GAN vocoder. The speaker encoder, based on the pre-trained RawNet3, extracts speaker-specific voice features. Furthermore, the proposed approach not only includes an English one-shot multi-speaker TTS but also introduces a Korean one-shot multi-speaker TTS. We evaluate naturalness and speaker similarity of the generated speech using objective and subjective metrics. In the subjective evaluation, the proposed Korean one-shot multi-speaker TTS obtained naturalness mean opinion score (NMOS) of 3.36 and similarity MOS (SMOS) of 3.16. The objective evaluation of the proposed English and Korean one-shot multi-speaker TTS showed a prediction MOS (P-MOS) of 2.54 and 3.74, respectively. These results indicate that the performance of our proposed model is improved over the baseline models in terms of both naturalness and speaker similarity.

• Journal of Information Processing Systems
• /
• v.9 no.1
• /
• pp.117-140
• /
• 2013

Hindi is the most widely spoken language in India, with more than 300 million speakers. As there is no separation between the characters of texts written in Hindi as there is in English, the Optical Character Recognition (OCR) systems developed for the Hindi language carry a very poor recognition rate. In this paper we propose an OCR for printed Hindi text in Devanagari script, using Artificial Neural Network (ANN), which improves its efficiency. One of the major reasons for the poor recognition rate is error in character segmentation. The presence of touching characters in the scanned documents further complicates the segmentation process, creating a major problem when designing an effective character segmentation technique. Preprocessing, character segmentation, feature extraction, and finally, classification and recognition are the major steps which are followed by a general OCR. The preprocessing tasks considered in the paper are conversion of gray scaled images to binary images, image rectification, and segmentation of the document's textual contents into paragraphs, lines, words, and then at the level of basic symbols. The basic symbols, obtained as the fundamental unit from the segmentation process, are recognized by the neural classifier. In this work, three feature extraction techniques-: histogram of projection based on mean distance, histogram of projection based on pixel value, and vertical zero crossing, have been used to improve the rate of recognition. These feature extraction techniques are powerful enough to extract features of even distorted characters/symbols. For development of the neural classifier, a back-propagation neural network with two hidden layers is used. The classifier is trained and tested for printed Hindi texts. A performance of approximately 90% correct recognition rate is achieved.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.5
• /
• pp.74-82
• /
• 1995

This paper is study on the voice dialing using HMM and post processing of the connected digits. HMM algorithm is widely used in the speech recognition with a good result. But, the maximum likelihood estimation of HMM(Hidden Markov Model) training in the speech recognition does not lead to values which maximize recognition rate. To solve the problem, we applied the post processing to segmental K-means procedure are in the recognition experiment. Korea connected digits are influenced by the prolongation more than English connected digits. To decrease the segmentation error in the level building algorithm some word models which can be produced by the prolongation are added. Some rules for the added models are applied to the recognition result and it is updated. The recognition system was implemented with DSP board having a TMS320C30 processor and IBM PC. The reference patterns were made by 3 male speakers in the noisy laboratory. The recognition experiment was performed for 21 sort of telephone number, 252 data. The recognition rate was $6\%$ in the speaker dependent, and $80.5\%$ in the speaker independent recognition test.

• Speech Sciences
• /
• v.7 no.3
• /
• pp.197-218
• /
• 2000

The vocal tract shapes were measured from tracheoesophageal speakers during the sustained phonation of five Korean vowels /u/, /o/, /a/, /e/, /i/ using magnetic resonance image(MRI). The subject's original vowel utterances with speech intelligibility and the synthesized vowels from MR images were analyzed. The results were as follows: (1) The vowels /a/, /e/, /i/ were perceived as the same sounds of actual subject's speech, but the vowels /o/ and /u/ were perceived as /$\partial$/ and strained /u/, respectively. (2) The synthesized vowels /a/ and /e/ from the MR images were perceived as the same sounds, but the vowels /u/, /o/, /i/ were perceived as different sounds. (3) The synthesized vowel by the expanded pharyngeal segment of 3 times in vowel /o/ was perceived as more natural than that of 2 times. The pharyngeal areas with varied sizes should be experimented to secure better speech production because the correct shapes of the vocal tract lead to distinct vowel production.

• Journal of the Korean association of regional geographers
• /
• v.13 no.6
• /
• pp.639-650
• /
• 2007

This paper deals with construction of a new administrative city in Naju. This kind of cities are commonly called as 'innovation city' in Korea. This is an experimental regional development program for non-Capital regions. State-owned corporations and public agencies are going to be relocated in the cities. Whether the employees will settle down in Naju with their families is the main research question in this paper. The result of the survey shows that more than half of the respondents are not going to settle down with their families. They are not optimistic about the present and future of the city either. In order to induce in-migration with families, educational environment need to be improved. To help their children to have more chance to talk with native English speakers is one of the examples. Facilities for the illed and the old are also necessary.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.353-358
• /
• 2021

This study provides the syntactic analysis for the agreement attraction by proposing three types of syntactic subject-verb agreement. Because subject-verb number agreement codifies the link between a predicate and its subject, it must be the purely syntactic processes of the head-to-head agreement or the feature percolation, where relevant agreement features percolate upward or downward through the hierarchical syntactic structure. The agreement errors are not affected by linear proximity or minimal interference, but instead are affected by the hierarchical relationship between an agreement target and a local attractor. The data in this paper includes the complex noun phrases with a modifier PP or a relative clause CP. Here, the [+PL] feature is suggested to be a local attractor for subject-verb agreement errors as a strong feature. Therefore, speakers tend to erroneously produce plural agreement for a singular subject in a main clause due to a plural NP in a modifier PP or plural agreement for a singular subject in a relative clause due to plural main subject.

• Phonetics and Speech Sciences
• /
• v.14 no.3
• /
• pp.11-25
• /
• 2022

From simultaneous recordings of the external photoglottography, intra-oral air pressure (P_io), airflow and acoustic data from four native Seoul Korean speakers (2 male and 2 female), we have found that the two fricatives are not significantly different in glottal opening peak and airflow peak height either word-initially or word-medially and that the duration of aspiration is significantly reduced in word-medial /s/, compared to those in word-initial /s/, not in /s'/. We have also found that the duration of a high P_io plateau is significantly longer in /s/ than in /s'/ both word-initially and word-medially and that airflow resistance (R=P_io/U) at the onset and offset of a P_io plateau and at the time of airflow peak height is significantly higher in /s'/ than in /s/ across the contexts. However, the differences in P_io peak and F0 are not significant. In addition, the transition time to reach airflow peak height from the offset of a P_io plateau is found to be significantly longer in /s/ than /s'/ in both word-initial and word-medial positions. No significant differences in glottal opening peak and airflow peak height confirm that /s/ is specified as [-spread glottis] like /s'/. As for the other significant differences, we propose that /s/ is [-tense], and /s'/ [+tense].

• Phonetics and Speech Sciences
• /
• v.11 no.1
• /
• pp.9-18
• /
• 2019

The purpose of this study is to determine whether the mean fundamental frequency (F0) changes depending on language, task, or language proficiency for Korean-English bilinguals. A total of forty-eight Korean-English speakers (28 balanced bilinguals and 20 Korean dominant bilinguals) participated in the study. Participants were asked to read aloud two types of tasks in English and Korean. For statistical analyses, the language ${\times}$ task two-way repeated ANOVAs were conducted within the balanced bilingual group first, and then group ${\times}$ language two-way mixed ANOVAs. The results showed that the females in both bilingual groups changed their mean F0 depending on the language they used and the tasks (p<.05), whereas no significant results were found in the males in either group under any conditions. The mean fundamental frequency in the Korean reading task was significantly higher than that in the English reading task for females in both balanced and Korean dominant bilingual groups. Thus, changes in mean F0 depending on language and context may reflect gender-specific characteristics, and females seem to be more sensitive to the socio-cultural standards that are imposed on them.