• 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.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.6
• /
• pp.277-281
• /
• 2006

This paper presents a method for the automatic recognition of pitch accents over syllables. The method that we propose is based on the time-delay recursive neural network (TDRNN). which is a neural network classifier with two different representation of dynamic context: the delayed input nodes allow the representation of an explicit trajectory F0(t) along time. while the recursive nodes provide long-term context information that reflects the characteristics of pitch accentuation in spoken English. We apply the TDRNN to pitch accent recognition in two forms: in the normal TDRNN. all of the prosodic features (pitch. energy, duration) are used as an entire set in a single TDRNN. while in the distributed TDRNN. the network consists of several TDRNNs each taking a single prosodic feature as the input. The final output of the distributed TDRNN is weighted sum of the output of individual TDRNN. We used the Boston Radio News Corpus (BRNC) for the experiments on the speaker-independent pitch accent recognition. π 1e experimental results show that the distributed TDRNN exhibits an average recognition accuracy of 83.64% over both pitch events and non-events.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2
• /
• pp.155-163
• /
• 2009

In text-to-speech systems, the conversion of text into prosodic parameters is necessarily composed of three steps. These are the placement of prosodic boundaries. the determination of segmental durations, and the specification of fundamental frequency contours. Prosodic boundaries. as the most important and basic parameter. affect the estimation of durations and fundamental frequency. Break prediction is an important step in text-to-speech systems as break indices (BIs) have a great influence on how to correctly represent prosodic phrase boundaries, However. an accurate prediction is difficult since BIs are often chosen according to the meaning of a sentence or the reading style of the speaker. In Japanese, the prediction of an accentual phrase boundary (APB) and major phrase boundary (MPB) is particularly difficult. Thus, this paper presents a method to complement the prediction errors of an APB and MPB. First, we define a subtle BI in which it is difficult to decide between an APB and MPB clearly as a variable break (VB), and an explicit BI as a fixed break (FB). The VB is chosen using the classification and regression tree, and multiple prosodic targets in relation to the pith and duration are then generated. Finally. unit-selection is conducted using multiple prosodic targets. In the MOS test result. the original speech scored a 4,99. while proposed method scored a 4.25 and conventional method scored a 4.01. The experimental results show that the proposed method improves the naturalness of synthesized speech.

• 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.

• Journal of Intelligence and Information Systems
• /
• v.26 no.3
• /
• pp.127-147
• /
• 2020

The world is suffering from numerous human and economic losses due to the novel coronavirus infection (COVID-19). The Korean government established a strategy to overcome the national infectious disease crisis through research and development. It is difficult to find distinctive features and changes in a specific R&D field when using the existing technical classification or science and technology standard classification. Recently, a few studies have been conducted to establish a classification system to provide information about the investment research areas of infectious diseases in Korea through a comparative analysis of Korea government-funded research projects. However, these studies did not provide the necessary information for establishing cooperative research strategies among countries in the infectious diseases, which is required as an execution plan to achieve the goals of national health security and fostering new growth industries. Therefore, it is inevitable to study information services based on the classification system and classification model for establishing a national collaborative R&D strategy. Seven classification - Diagnosis_biomarker, Drug_discovery, Epidemiology, Evaluation_validation, Mechanism_signaling pathway, Prediction, and Vaccine_therapeutic antibody - systems were derived through reviewing infectious diseases-related national-funded research projects of South Korea. A classification system model was trained by combining Scopus data with a bidirectional RNN model. The classification performance of the final model secured robustness with an accuracy of over 90%. In order to conduct the empirical study, an infectious disease classification system was applied to the coronavirus-related research and development projects of major countries such as the STAR Metrics (National Institutes of Health) and NSF (National Science Foundation) of the United States(US), the CORDIS (Community Research & Development Information Service)of the European Union(EU), and the KAKEN (Database of Grants-in-Aid for Scientific Research) of Japan. It can be seen that the research and development trends of infectious diseases (coronavirus) in major countries are mostly concentrated in the prediction that deals with predicting success for clinical trials at the new drug development stage or predicting toxicity that causes side effects. The intriguing result is that for all of these nations, the portion of national investment in the vaccine_therapeutic antibody, which is recognized as an area of research and development aimed at the development of vaccines and treatments, was also very small (5.1%). It indirectly explained the reason of the poor development of vaccines and treatments. Based on the result of examining the investment status of coronavirus-related research projects through comparative analysis by country, it was found that the US and Japan are relatively evenly investing in all infectious diseases-related research areas, while Europe has relatively large investments in specific research areas such as diagnosis_biomarker. Moreover, the information on major coronavirus-related research organizations in major countries was provided by the classification system, thereby allowing establishing an international collaborative R&D projects.