• The KIPS Transactions:PartB
• /
• v.10B no.2
• /
• pp.169-178
• /
• 2003

This study outlines two viewpoints the classification of phone likely unit (PLU) which is the foundation of korean large vocabulary speech recognition, and the effectiveness of Chiljongseong (7 Final Consonants) and Paljogseong (8 Final Consonants) of the korean language. The phone likely classifies the phoneme phonetically according to the location of and method of articulation, and about 50 phone-likely units are utilized in korean speech recognition. In this study auditory phonetical knowledge was applied to the classification of phone likely unit to present 45 phone likely unit. The vowels 'ㅔ, ㅐ'were classified as phone-likely of (ee) ; 'ㅒ, ㅖ' as [ye] ; and 'ㅚ, ㅙ, ㅞ' as [we]. Secondly, the Chiljongseong System of the draft for unified spelling system which is currently in use and the Paljongseonggajokyong of Korean script haerye were illustrated. The question on whether the phonetic value on 'ㄷ' and 'ㅅ' among the phonemes used in the final consonant of the korean fan guage is the same has been argued in the academic world for a long time. In this study, the transition stages of Korean consonants were investigated, and Ciljonseeng and Paljongseonggajokyong were utilized in speech recognition, and its effectiveness was verified. The experiment was divided into isolated word recognition and speech recognition, and in order to conduct the experiment PBW452 was used to test the isolated word recognition. The experiment was conducted on about 50 men and women - divided into 5 groups - and they vocalized 50 words each. As for the continuous speech recognition experiment to be utilized in the materialized stock exchange system, the sentence corpus of 71 stock exchange sentences and speech corpus vocalizing the sentences were collected and used 5 men and women each vocalized a sentence twice. As the result of the experiment, when the Paljongseonggajokyong was used as the consonant, the recognition performance elevated by an average of about 1.45% : and when phone likely unit with Paljongseonggajokyong and auditory phonetic applied simultaneously, was applied, the rate of recognition increased by an average of 1.5% to 2.02%. In the continuous speech recognition experiment, the recognition performance elevated by an average of about 1% to 2% than when the existing 49 or 56 phone likely units were utilized.

• Phonetics and Speech Sciences
• /
• v.7 no.3
• /
• pp.45-53
• /
• 2015

The current study investigated CVCC syllables in spontaneous American English speech to find out whether such syllables are produced as phonological units with a string of segments, showing a hierarchical structure. Transcribed data from the Buckeye Speech Corpus was used for the analysis in this study. The result of the current study showed that the constituents within a CVCC syllable as a phonological unit may have phonetic variations (namely, the final coda may undergo deletion). First, voiceless alveolar stops were the most frequently deleted when they occurred as the second final coda consonants of a CVCC syllable; this deletion may be an intermediate process on the way from the abstract form CVCC (with the rime VCC) to the actual pronunciation CVC (with the rime VC), a production strategy employed by some individual speakers. Second, in the internal structure of the rime, the proportion of deletion of the final coda consonant depended on the frequency of the word rather than on the position of postvocalic consonants on the sonority hierarchy. Finally, the segment following the consonant cluster proved to have an effect on the reduction of that cluster; more precisely, the following contrast was observed between obstruents and non-obstruents, reflecting the effect of sonority: when the segment following the consonant cluster was an obstruent, the proportion of deletion of the final coda consonant was increased. Among these results, the effect of word frequency played a critical role for promoting the deletion of the second coda consonant for clusters in CVCC syllables in spontaneous speech. The current study implies that the structure of syllables as phonological units can vary depending on individual speakers' lexical representation.

• Phonetics and Speech Sciences
• /
• v.7 no.3
• /
• pp.139-149
• /
• 2015

This paper aims to analyze pronunciation variations of loanwords produced by Korean and improve the performance of pronunciation modeling of loanwords in Korean by using syllable-based segmentation and phonological knowledge. The loanword text corpus used for our experiment consists of 14.5k words extracted from the frequently used words in set-top box, music, and point-of-interest (POI) domains. At first, pronunciations of loanwords in Korean are obtained by manual transcriptions, which are used as target pronunciations. The target pronunciations are compared with the standard pronunciation using confusion matrices for analysis of pronunciation variation patterns of loanwords. Based on the confusion matrices, three salient pronunciation variations of loanwords are identified such as tensification of fricative [s] and derounding of rounded vowel [ɥi] and [$w{\varepsilon}$]. In addition, a syllable-based segmentation method considering phonological knowledge is proposed for loanword pronunciation modeling. Performance of the baseline and the proposed method is measured using phone error rate (PER)/word error rate (WER) and F-score at various context spans. Experimental results show that the proposed method outperforms the baseline. We also observe that performance degrades when training and test sets come from different domains, which implies that loanword pronunciations are influenced by data domains. It is noteworthy that pronunciation modeling for loanwords is enhanced by reflecting phonological knowledge. The loanword pronunciation modeling in Korean proposed in this paper can be used for automatic speech recognition of application interface such as navigation systems and set-top boxes and for computer-assisted pronunciation training for Korean learners of English.

• Phonetics and Speech Sciences
• /
• v.15 no.1
• /
• pp.9-17
• /
• 2023

The purpose of this study is to determine how the Korean diphthong /jo/ shows phonetic variation in various linguistic environments. The pronunciation of /jo/ is discussed, focusing on the relationship between phonetic variation and the distribution range of vowels. The location in a word (monosyllable, word-initial, word-medial, word-final) and word class (content word, function word) were analyzed using the speech of 10 female speakers of the Seoul Corpus. As a result of determining the frequency of appearance of /jo/ in each environment, the pronunciation type and word class were affected by the location in a word. Frequent phonetic reduction was observed in the function word /jo/ in the acoustic analysis. The word class did not change the average phonetic values of /jo/, but changed the distribution of individual tokens. These results indicate that the linguistic environment affects the phonetic distribution of vowels.

• Phonetics and Speech Sciences
• /
• v.12 no.2
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.

• Phonetics and Speech Sciences
• /
• v.15 no.1
• /
• pp.25-34
• /
• 2023

This study examined the utility of the acoustic features of vowels as cues for the place of articulation of Korean nasal consonants. In the acoustic analysis, spectral and temporal parameters were measured at the 25%, 50%, and 75% time points in the vowels neighboring nasal consonants in samples extracted from a spontaneous Korean speech corpus. Using these measurements, linear discriminant analyses were performed and classification accuracies for the nasal place of articulation were estimated. The analyses were applied separately for vowels following and preceding a nasal consonant to compare the effects of progressive and regressive coarticulation in terms of place of articulation. The classification accuracies ranged between approximately 50% and 60%, implying that acoustic measurements of vowel intervals alone are not sufficient to predict or classify the place of articulation of adjacent nasal consonants. However, given that these results were obtained for measurements at the temporal midpoint of vowels, where they are expected to be the least influenced by coarticulation, the present results also suggest the potential of utilizing acoustic measurements of vowels to improve the recognition accuracy of nasal place. Moreover, the classification accuracy for nasal place was higher for vowels preceding the nasal sounds, suggesting the possibility of higher anticipatory coarticulation reflecting the nasal place.

• Phonetics and Speech Sciences
• /
• v.14 no.1
• /
• pp.37-47
• /
• 2022

The purpose of the this study is to observe the effects of healthy adults' age on temporal features of speech and identify which could differentiate older and young adults. We examined speech rates(i.e., overall speaking rate, articulation rate), occurrence of pause, and duration of pause per utterance by utilizing the National Institute of Korean Language's open corpus. We selected a total of 30 healthy adults (10 young, 10 middle-aged, and 10 older adults) in this study. There were significant differences among the groups in the overall speaking rate, articulation rate, total occurrence of pause, the occurrence of pause between syntactic words, total duration of pause, and duration of pause between syntactic words. The older and middle-aged adults showed slower speech rates and longer and more frequent pause than young adults. But there were no significant differences among the three groups in terms of pause within syntactic word. The overall speaking rate significantly differentiated older adults from young adults. These findings suggested that the effect of speakers' age was reflected in gradual changes in the temporal features of their speech.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Phonetics and Speech Sciences
• /
• v.12 no.3
• /
• pp.33-40
• /
• 2020

The purpose of this research was to find out the relation between Korean learners' English proficiency and the ratio of the length of the stressed vowel in a monosyllabic noun to that of the unstressed vowel in an article of the noun phrases (e.g., "a cup", "the bus", etcs.). Generally, the vowels in monosyllabic content words are phonetically more prominent than the ones in monosyllabic function words as the former have phrasal stress, making the vowels in content words longer in length, higher in pitch, and louder in amplitude. This study, based on the speech samples from Korean-Spoken English Corpus (K-SEC) and Rated Korean-Spoken English Corpus (Rated K-SEC), examined 879 English noun phrases, which are composed of an article and a monosyllabic noun, from sentences which are rated on 4 levels of proficiency. The lengths of the vowels in these 879 target NPs were measured and the ratio of the vowel lengths in nouns to those in articles was calculated. It turned out that the higher the proficiency level, the greater the mean ratio of the vowels in nouns to the vowels in articles, confirming the research's hypothesis. This research thus concluded that for the Korean English learners, the higher the English proficiency level, the better they could produce the stressed and unstressed vowels with more conspicuous length differences between them.