• The Journal of the Acoustical Society of Korea
• /
• v.15 no.6
• /
• pp.5-11
• /
• 1996

This paper is the study for recognizing single vowels of speaker-independent, and we suppose a method of speech recognition using VQ(Vector Quantization)/NN(Neural Network). This method makes a VQ codebook, which is used for obtaining the observation sequence, and then claculates the probability value by comparing each codeword with the data, finally uses these probability values for the input value of the neural network. Korean signle vowels are selected for our recognition experiment, and ten male speakers pronounced eight single vowels ten times. We compare the performance of our method with those of fuzzy VQ/HMM and conventional VQ/NN According to the experiment result, the recognition rate by VQ/NN is 92.3%, by VQ/HMM using fuzzy is 93.8% and by VQ/NN using fuzzy is 95.7%. Therefore, it is shown that recognition rate of speech recognition by fuzzy VQ/NN is better than those of fuzzy VQ/HMM and conventional VQ/HMM because of its excellent learning ability.

• Journal of Engineering Education Research
• /
• v.20 no.3
• /
• pp.3-12
• /
• 2017

This research points out the necessity of customized communication education for engineering students in the era of the Fourth Industrial Revolution. This paper also points out such problems of current communication education as presentation and discussion-focused 'Public Speech' exercises, absence of interests about social issues, and lack of interactive communication learning. In general, as the characteristics of their major education, engineering students are not aggressive in self-questioning and active communication rather than their sensitive reaction to the changes of the new era. Considering these characteristics of engineering students, this research emphasizes that future communication education should be deployed from the major-focused thinking to the development of convergent thinking, from the problem-solving to the problem-finding, and from the contentious thinking to the cooperative thinking. In addition, as a class design reflecting future trends, this research emphasizes, firstly the development of cooperative communication education model, secondly active utilization of SMART technology, and lastly the importance of customized-coaching for each student considering their own characteristics and requirements.

• ETRI Journal
• /
• v.36 no.5
• /
• pp.704-713
• /
• 2014

Most natural language processing tasks depend on the outputs of some other tasks. Thus, they involve other tasks as subtasks. The main problem of this type of pipelined model is that the optimality of the subtasks that are trained with their own data is not guaranteed in the final target task, since the subtasks are not optimized with respect to the target task. As a solution to this problem, this paper proposes a consolidation of subtasks for a target task ($CST^2$). In $CST^2$, all parameters of a target task and its subtasks are optimized to fulfill the objective of the target task. $CST^2$ finds such optimized parameters through a backpropagation algorithm. In experiments in which text chunking is a target task and part-of-speech tagging is its subtask, $CST^2$ outperforms a traditional pipelined text chunker. The experimental results prove the effectiveness of optimizing subtasks with respect to the target task.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.5
• /
• pp.225-237
• /
• 2022

The term "artificial intelligence" is considered a component of sophisticated technological developments, and several intelligent tools have been developed to assist organizations and entrepreneurs in making business decisions. Artificial intelligence (AI) is defined as the concept of transforming inanimate objects into intelligent beings that can reason in the same way that humans do. Computer systems can imitate a variety of human intelligence activities, including learning, reasoning, problem-solving, speech recognition, and planning. This study's objective is to provide responses to the questions: Which factors should be taken into account while deciding whether or not to use AI applications? What role do these elements have in AI application adoption? However, this study proposes a framework to explore the significance and relation of success factors to AI adoption based on the technology-organization-environment model. Ten critical factors related to AI adoption are identified. The framework is empirically tested with data collected by mail surveying organizations in Vietnam. Structural Equation Modeling is applied to analyze the data. The results indicate that Technical compatibility, Relative advantage, Technical complexity, Technical capability, Managerial capability, Organizational readiness, Government involvement, Market uncertainty, and Vendor partnership are significantly related to AI applications adoption.

• Journal of KIISE
• /
• v.45 no.2
• /
• pp.157-164
• /
• 2018

Due to the revolutionary advances in deep learning, performance of pattern recognition has increased significantly in many applications like speech recognition and image recognition, and some systems outperform human-level intelligence in specific domains. Unlike pattern recognition, in this paper, we focus on generating Korean sentences based on a few Korean sentences. We apply variational recurrent auto-encoder (VRAE) and modify the model considering some characteristics of Korean sentences. To reduce the number of words in the model, we apply a word spacing model. Also, there are many Korean sentences which have the same meaning but different word order, even without subjects or objects; therefore we change the unidirectional encoder of VRAE into a bidirectional encoder. In addition, we apply an interpolation method on the encoded vectors from the given sentences, so that we can generate new sentences which are similar to the given sentences. In experiments, we confirm that our proposed method generates better sentences which are semantically more similar to the given sentences.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.1
• /
• pp.77-92
• /
• 2020

When automatic speech recognition (ASR) is provided as a cloud service, it is easy to collect voice and application domain data from users. Harnessing these data will facilitate the provision of more personalized services. In this paper, we demonstrate our transfer learning-based knowledge service that built with the user-generated data collected through our novel system that deliveries personalized ASR service. First, we discuss the motivation, challenges, and prospects of building up such a knowledge-based service-oriented system. Second, we present a Quadruple Transfer Learning (QTL) method that can learn a classification model from a source domain and transfer it to a target domain. Third, we provide an overview architecture of our novel system that collects voice data from mobile users, labels the data via crowdsourcing, utilises these collected user-generated data to train different machine learning models, and delivers the personalised real-time cloud services. Finally, we use the E-Book data collected from our system to train classification models and apply them in the smart TV domain, and the experimental results show that our QTL method is effective in two classification tasks, which confirms that the knowledge transfer provides a value-added service for the upper-layer mobile applications in different domains.

• The Journal of Korean Association of Computer Education
• /
• v.14 no.1
• /
• pp.137-146
• /
• 2011

Edward de Bono's six thinking hats is effective to increase the excellence of the thinking and cultivate divergent thinking. In particular, this method is effective in finding a reasonable solution through analyzing issue from a variety of perspective. In this paper, we developed the system for an effective debate learning that student's own ideas based on six thinking hats are shared and expressed in speech and images through Android devices sensors. We analyzed the tools and guidelines by making design structural model for system design. The developed system was evaluated by the teacher through demonstration and practice and we analyzed through evaluation results the effectiveness and the improvement of the system. The evaluation results were analysed that the developed system is more effective improve motivation and debate ability than web based debate learning system.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.252-255
• /
• 2018

화행 분석이란 자연어 발화를 통해 나타나는 화자의 의도를 파악하는 것을 말하며, 슬롯 필링이란 자연어 발화에서 도메인에 맞는 정보를 추출하기 위해 미리 정의되어진 슬롯에 대한 값을 찾는 것을 말한다. 최근 화행 분석과 슬롯 필링 연구는 딥 러닝 기반의 공동 학습을 이용하는 연구가 많이 이루어지고 있고 본 논문에서는 한국어 특허상담 도메인 대화 말뭉치를 이용하여 공동 학습 모델을 구축하고 개별적인 모델과 성능을 비교한다. 또한 추가적으로 공동 학습 모델에 주의집중 메커니즘을 적용하여 성능이 향상됨을 보인다. 최종적으로 주의집중 메커니즘 기반의 공동 학습 모델이 기준 모델과 비교하여 화행 분류와 슬롯 필링 성능이 각각 3.35%p, 0.54%p 향상되어 85.41%, 80.94%의 성능을 얻었다.

• ETRI Journal
• /
• v.45 no.4
• /
• pp.678-689
• /
• 2023

This paper describes a novel end-to-end deep generative model-based speaker recognition system using prosodic features. The usefulness of variational autoencoders (VAE) in learning the speaker-specific prosody representations for the speaker recognition task is examined herein for the first time. The speech signal is first automatically segmented into syllable-like units using vowel onset points (VOP) and energy valleys. Prosodic features, such as the dynamics of duration, energy, and fundamental frequency (F₀), are then extracted at the syllable level and used to train/adapt a speaker-dependent VAE from a universal VAE. The initial comparative studies on VAEs and traditional autoencoders (AE) suggest that the former can efficiently learn speaker representations. Investigations on the impact of gender information in speaker recognition also point out that gender-dependent impostor banks lead to higher accuracies. Finally, the evaluation on the NIST SRE 2010 dataset demonstrates the usefulness of the proposed approach for speaker recognition.

• The Korean Journal of Applied Statistics
• /
• v.36 no.2
• /
• pp.101-114
• /
• 2023

The data augmentation technique is effectively used to solve the problem of overfitting the model by allowing the training dataset to be viewed from various perspectives. In addition to image augmentation techniques such as rotation, cropping, horizontal flip, and vertical flip, occlusion-based data augmentation methods such as Cutmix and Cutout have been proposed. For models based on speech data, it is possible to use an occlusion-based data-based augmentation technique after converting a 1D speech signal into a 2D spectrogram. In particular, SpecAugment is an occlusion-based augmentation technique for speech spectrograms. In this study, we intend to compare and study data augmentation techniques that can be used in the problem of false-voice detection. Using data from the ASVspoof2017 and ASVspoof2019 competitions held to detect fake audio, a dataset applied with Cutout, Cutmix, and SpecAugment, an occlusion-based data augmentation method, was trained through an LCNN model. All three augmentation techniques, Cutout, Cutmix, and SpecAugment, generally improved the performance of the model. In ASVspoof2017, Cutmix, in ASVspoof2019 LA, Mixup, and in ASVspoof2019 PA, SpecAugment showed the best performance. In addition, increasing the number of masks for SpecAugment helps to improve performance. In conclusion, it is understood that the appropriate augmentation technique differs depending on the situation and data.