• MALSORI
• /
• no.46
• /
• pp.145-157
• /
• 2003

In this paper, we'd like to briefly introduce Korean common speech database, which project has been started to construct a large scaled speech database since 2002. The project aims at supporting the R&D environment of the speech technology for industries. It encourages domestic speech industries and activates speech technology domestic market. In the first year, the resulting common speech database consists of 25 kinds of databases considering various recording conditions such as telephone, PC, VoIP etc. The speech database will be widely used for speech recognition, speech synthesis, and speaker identification. On the other hand, although the database was originally corrected by manual, still it retains unknown errors and human errors. So, in order to minimize the errors in the database, we tried to find the errors based on the recognition errors and classify several kinds of errors. To be more effective than typical recognition technique, we will develop the automatic error detection method. In the future, we will try to construct new databases reflecting the needs of companies and universities.

• Korean Journal of Cognitive Science
• /
• v.17 no.2
• /
• pp.139-150
• /
• 2006

Speech recognition errors cause fatal results in a spoken dialogue system. When a system can not determine the speech-act of u utterance due to speech recognition errors, a dialogue system has a difficulty in continuing conversation. In this paper, we propose strategies for sub-dialogue generation by inferring the speech-act of an utterance with patterns of recognition errors on the field of form-filling dialogue. We used the proposed method on a plan-based dialogue model, corrected 27% of incomplete tasks, and acquired overall 89% of task completion rate.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.267-286
• /
• 2023

The conversational agents such as AI speakers utilize voice conversation for human-computer interaction. Voice recognition errors often occur in conversational situations. Recognition errors in user utterance records can be categorized into two types. The first type is misrecognition errors, where the agent fails to recognize the user's speech entirely. The second type is misinterpretation errors, where the user's speech is recognized and services are provided, but the interpretation differs from the user's intention. Among these, misinterpretation errors require separate error detection as they are recorded as successful service interactions. In this study, various text separation methods were applied to detect misinterpretation. For each of these text separation methods, the similarity of consecutive speech pairs using word embedding and document embedding techniques, which convert words and documents into vectors. This approach goes beyond simple word-based similarity calculation to explore a new method for detecting misinterpretation errors. The research method involved utilizing real user utterance records to train and develop a detection model by applying patterns of misinterpretation error causes. The results revealed that the most significant analysis result was obtained through initial consonant extraction for detecting misinterpretation errors caused by the use of unregistered neologisms. Through comparison with other separation methods, different error types could be observed. This study has two main implications. First, for misinterpretation errors that are difficult to detect due to lack of recognition, the study proposed diverse text separation methods and found a novel method that improved performance remarkably. Second, if this is applied to conversational agents or voice recognition services requiring neologism detection, patterns of errors occurring from the voice recognition stage can be specified. The study proposed and verified that even if not categorized as errors, services can be provided according to user-desired results.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.1-7
• /
• 2021

Recently, since most of the research on correcting speech recognition errors is based on English, there is not enough research on Korean speech recognition. Compared to English speech recognition, however, Korean speech recognition has many errors due to the linguistic characteristics of Korean language, such as Korean Fortis and Korean Liaison, thus research on Korean speech recognition is needed. Furthermore, earlier works primarily focused on editorial distance algorithms and syllable restoration rules, making it difficult to correct the error types of Korean Fortis and Korean Liaison. In this paper, we propose a context-sensitive post-processing model of speech recognition using a LSTM-based sequence-to-sequence model and Bahdanau attention mechanism to correct Korean speech recognition errors caused by the pronunciation. Experiments showed that by using the model, the speech recognition performance was improved from 64% to 77% for Fortis, 74% to 90% for Liaison, and from 69% to 84% for average recognition than before. Based on the results, it seems possible to apply the proposed model to real-world applications based on speech recognition.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.5
• /
• pp.430-437
• /
• 2021

In this paper, we propose a robust face recognition method against soft errors with a deep convolutional generative adversarial network(DCGAN) based compensation method by a cross-layer approach. When soft-errors occur in block data of JPEG files, these blocks can be decoded inappropriately. In previous results, these blocks have been replaced using a mean face, thereby improving recognition ratio to a certain degree. This paper uses a DCGAN-based compensation approach to extend the previous results. When soft errors are detected in an embedded system layer using parity bit checkers, they are compensated in the application layer using compensated block data by a DCGAN-based compensation method. Regarding soft errors and block data loss in facial images, a DCGAN architecture is redesigned to compensate for the block data loss. Simulation results show that the proposed method effectively compensates for performance degradation due to soft errors.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.537-542
• /
• 2021

This paper improves the possibility of recognizing fast-moving objects through the YOLO (You Only Look Once) deep learning recognition model in an application environment for object recognition in images. The purpose was to study the method of collecting semantic data through processing. In the recognition model, the moving object recognition error was identified as unrecognized because of the difference between the frame rate of the camera and the moving speed of the object and a misrecognition due to the existence of a similar object in an environment adjacent to the object. To minimize the recognition errors by compensating for errors, such as unrecognized and misrecognized objects through the proposed data collection method, and applying vision processing technology for the causes of errors that may occur in images acquired for sports (tennis games) that can represent real similar environments. The effectiveness of effective secondary data collection was improved by research on methods and processing structures. Therefore, by applying the data collection method proposed in this study, ordinary people can collect and manage data to improve their health and athletic performance in the sports and health industry through the simple shooting of a smart-phone camera.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.1E
• /
• pp.57-63
• /
• 1997

In this paper, we investigated performance of a vocabulary-independent speech recognizer with speaker adaptation. The vocabulary-independent speech recognizer does not require task-oriented speech databases to estimate HMM parameters, but adapts the parameters recursively by using input speech and recognition results. The recognizer has the advantage that it relieves efforts to record the speech databases and can be easily adapted to a new task and a new speaker with different recognition vocabulary without losing recognition accuracies. Experimental results showed that the vocabulary-independent speech recognizer with supervised offline speaker adaptation reduced 40% of recognition errors when 80 words from the same vocabulary as test data were used as adaptation data. The recognizer with unsupervised online speaker adaptation reduced abut 43% of recognition errors. This performance is comparable to that of a speaker-independent speech recognizer trained by a task-oriented speech database.

• 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 Computing Science and Engineering
• /
• v.2 no.1
• /
• pp.1-25
• /
• 2008

Spoken dialog tasks incur many errors including speech recognition errors, understanding errors, and even dialog management errors. These errors create a big gap between the user's intention and the system's understanding, which eventually results in a misinterpretation. To fill in the gap, people in human-to-human dialogs try to clarify the major causes of the misunderstanding to selectively correct them. This paper presents a method of clarification techniques to human-to-machine spoken dialog systems. We viewed the clarification dialog as a two-step problem-Belief confirmation and Clarification strategy establishment. To confirm the belief, we organized the clarification process into three systematic phases. In the belief confirmation phase, we consider the overall dialog system's processes including speech recognition, language understanding and semantic slot and value pairs for clarification dialog management. A clarification expert is developed for establishing clarification dialog strategy. In addition, we proposed a new design of plugging clarification dialog module in a given expert based dialog system. The experiment results demonstrate that the error verifiers effectively catch the word and utterance-level semantic errors and the clarification experts actually increase the dialog success rate and the dialog efficiency.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.3E
• /
• pp.31-36
• /
• 1996

In this paper, a two-stage recognition approach based on error pattern hypotheses is proposed to reduce errors of a connected digit recognizer. In the approach, a conventional recognizer is first used to produce N-best candidate strings, and then error patterns are hypothesized by examining the candidate strings. For substitution error pattern hypotheses, error-pattern-dependent classifiers having more discriminative power than the first-stage classifier are used ; and for insertion and deletion errors, word duration and energy contour information are exploited are exploited to discriminated confusing pairs. Simulation results showed that the proposed approach achieves 15% decrease in word error rate for speaker-independent Korean connected digit recognition when a hidden Markov model-based recognizer is used for the first-stage classifier.