• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.149-152
• /
• 2003

In this paper, data-driven temporal filter methods[1] are investigated for robust feature extraction. A principal component analysis technique is applied to the time trajectories of feature sequences of training speech data to get appropriate temporal filters. We did recognition experiments on the Korean connected digit telephone speech database released by SITEC, with data-driven temporal filters. Experimental results are discussed with our findings.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.6
• /
• pp.647-654
• /
• 2022

Although the fields in which is utilized children's speech recognition is on the rise, the lack of quality data is an obstacle to improving children's speech recognition performance. This paper proposes a new method for improving children's speech recognition performance by additionally using adult speech data. The proposed method is a transformer based domain adversarial training using dynamically weighted loss to effectively address the data imbalance gap between age that grows as the amount of adult training data increases. Specifically, the degree of class imbalance in the mini-batch during training was quantified, and the loss function was defined and used so that the smaller the data, the greater the weight. Experiments validate the utility of proposed domain adversarial training following asymmetry between adults and children training data. Experiments show that the proposed method has higher children's speech recognition performance than traditional domain adversarial training method under all conditions in which asymmetry between age occurs in the training data.

• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.61-64
• /
• 2003

In this paper, we introduce a new description method of annotation information of speech database. As one of structured description methods, XML based description which has been standardized by W3C will be applied to represent metadata of speech database. It will be continuously revised through the speech technology standard forum during this year

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2E
• /
• pp.66-71
• /
• 2009

In practical applications of speech recognition, one of the fundamental questions might be on the number of training data that should be provided for a specific task. Though plenty of training data would undoubtedly enhance the system performance, we are then faced with the problem of heavy cost. Therefore, it is of crucial importance to determine the least number of training data that will afford a certain level of accuracy. For this purpose, we investigate the effect of the number of training data on the speaker-independent speech recognition of isolated words by using FVQ/HMM. The result showed that the error rate is roughly inversely proportional to the number of training data and grows linearly with the vocabulary size.

• Phonetics and Speech Sciences
• /
• v.1 no.3
• /
• pp.3-18
• /
• 2009

Do loanword properties emerge in the acquisition of a foreign language and if so, how? Classic studies in adult language learning assumed loanword properties that range from near-ceiling to near-chance level of appearance depending on speech proficiency. The present research argues that such variations reflect different phonological types, rather than speech proficiency. To investigate the difference between learner speech and loanword phonology, the current research analyzes the speech data from five different proficiency levels of 92 Korean speakers who read 19 pairs of English words and sentences that contained loanwords. The experimental method is primarily an acoustical one, by which the phonological cause in the loanwords (e.g., the insertion of [$\Box$] at the end of the word stamp) would be attested to appear in learner speech, in comparison with native speech from 11 English speakers and 11 Korean speakers. The data investigated for the research are of segment deletion, insertion, substitution, and alternation in both learner speech and the native speech. The results indicate that learner speech does not present the loanword properties in many cases, but depends on the types of phonological causes. The relatively easy acquisition of target pronunciation is evidenced in the cases of segment deletion, insertion, substitution, and alternation, except when the loanword property involves the successful command of the target phonology such as the de-aspiration of [p] in apple. Such a case of difficult learning draws a sharp distinction from the cases of easy learning in the development of learner speech, particularly beyond the intermediate level of proficiency. Overall, learner speech departs from loanword phonology and develops toward the native speech value, depending on phonological contrasts in the native and foreign languages.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.15 no.4
• /
• pp.300-304
• /
• 2015

A low bit rate speech coder based on the non-uniform sampling technique is proposed. The non-uniform sampling technique is based on the detection of inflection points (IP). A speech block is processed by the IP detector, and the detected IP pattern is compared with entries of the IP database. The address of the closest member of the database is transmitted with the energy of the speech block. In the receiver, the decoder reconstructs the speech block using the received address and the energy information of the block. As results, the coder shows fixed data rate contrary to the existing speech coders based on the non-uniform sampling. Through computer simulation, the usefulness of the proposed technique is shown. The SNR performance of the proposed method is approximately 5.27 dB with the data rate of 1.5 kbps.

• Proceedings of the KIEE Conference
• /
• 1987.07b
• /
• pp.1583-1586
• /
• 1987

In the study for effective speech control we designed a personal computer system with A/D converter in which the speech signal is transformed by digital data displayed graphically on the moniter and with a D/A converter in which the digital data is transformed into speech signal which people can hear. We analyzed the character of the speech signal produced by the system. We designed the adaptive noise cancel algorithm so that noise and Interference are cancelled whenever the speech signal is recognized by the computer system. This is a basic system for artificial Intelligence.

• Asia Pacific Journal of Corpus Research
• /
• v.2 no.2
• /
• pp.1-15
• /
• 2021

The study aims to explore the use of metadiscourse in the Bank Negara Malaysia Governor's speeches based on Hyland's Interpersonal Model of Metadiscourse. The corpus data consist of 343 speech texts, which were extracted from the Malaysian Corpus of Financial English (MacFE), amounting to 688,778 tokens. Adopting both quantitative and qualitative approaches to data analysis the study investigates (1) the overall use of metadiscourse in the Bank Negara Governor's speech texts and (2) the functions of the most prominent metadiscourse resources used and their functions in the speech texts. The findings reveal that the Governor's speech texts to be interactional rather than interactive, revealing a rich distribution of interactional metadiscourse resources, namely engagement markers, self-mention, hedges, boosters and attitude markers throughout the texts. The interactional metadiscourse resources function to establish speaker-audience engagement and alignment of views, as well as to express degree of uncertainty and certainty and attitudes. The study concludes that the speech texts are not merely informational or propositional, but rather interpersonal.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.48-58
• /
• 2024

This paper presents the development of language tutoring systems for nonnative speakers by leveraging advanced end-to-end automatic speech recognition (ASR) and proficiency evaluation. Given the frequent errors in non-native speech, high-performance spontaneous speech recognition must be applied. Our systems accurately evaluate pronunciation and speaking fluency and provide feedback on errors by relying on precise transcriptions. End-to-end ASR is implemented and enhanced by using diverse non-native speaker speech data for model training. For performance enhancement, we combine semisupervised and transfer learning techniques using labeled and unlabeled speech data. Automatic proficiency evaluation is performed by a model trained to maximize the statistical correlation between the fluency score manually determined by a human expert and a calculated fluency score. We developed an English tutoring system for Korean elementary students called EBS AI Peng-Talk and a Korean tutoring system for foreigners called KSI Korean AI Tutor. Both systems were deployed by South Korean government agencies.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.3
• /
• pp.451-458
• /
• 2001

In this paper, we propose speech control system for a various control device in the car with real time control speech. A real time speech control system is detected start-end points from speech data processing by A/D conversion, and recognize by one pass dynamic programming method. The results displays a monitor, and transports control data to control interfaces. The HMM model is modeled by a continuous control speech consists of control speech and digit speech for controlling of a various control device in the car The recognition rates is an average 97.3% in case of word & control speech, and is an average 96.3% in case of digit speech.