• MALSORI
• /
• v.68
• /
• pp.75-93
• /
• 2008

An example-based dialogue system tries to utilize many pairs which are stored in a dialogue database. The most important part of the example-based dialogue system is to find the most similar utterance to user's input utterance. Modality, which is characterized as conveying the speaker's involvement in the propositional content of a given utterance, is one of the core sentence features. For example, the sentence "I want to go to school." has a modality of hope. In this paper, we have proposed a modality classification system which can predict sentence modality in order to improve the performance of example-based dialogue systems. We also define a modality tag set for a dialogue system, and validate this tag set using a rule-based modality classification system. Experimental results show that our modality tag set and modality classification system improve the performance of an example-based dialogue system.

• Journal of KIISE
• /
• v.43 no.1
• /
• pp.80-86
• /
• 2016

A speech-act is a behavior intended by users in an utterance. Speech-act classification is important in a dialogue system. The machine learning and rule-based methods have mainly been used for speech-act classification. In this paper, we propose a speech-act classification method based on the combination of support vector machine (SVM) and transformation-based learning (TBL). The user's utterance is first classified by SVM that is preferentially applied to categories with a low utterance rate in training data. Next, when an utterance has negative scores throughout the whole of the categories, the utterance is applied to the correction phase by rules. The results from our method were higher performance over the baseline system long with error-reduction.

• ETRI Journal
• /
• v.36 no.5
• /
• pp.714-720
• /
• 2014

In this paper, we propose the use of data-driven probabilistic utterance-level decision logic to improve Weighted Finite State Transducer (WFST)-based endpoint detection. In general, endpoint detection is dealt with using two cascaded decision processes. The first process is frame-level speech/non-speech classification based on statistical hypothesis testing, and the second process is a heuristic-knowledge-based utterance-level speech boundary decision. To handle these two processes within a unified framework, we propose a WFST-based approach. However, a WFST-based approach has the same limitations as conventional approaches in that the utterance-level decision is based on heuristic knowledge and the decision parameters are tuned sequentially. Therefore, to obtain decision knowledge from a speech corpus and optimize the parameters at the same time, we propose the use of data-driven probabilistic utterance-level decision logic. The proposed method reduces the average detection failure rate by about 14% for various noisy-speech corpora collected for an endpoint detection evaluation.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.40-43
• /
• 2018

대화 발화 예측(Next Utterance Classification)은 Multi-turn 대화에서 마지막에 올 발화를 정답 후보들 중에서 예측을 하는 연구이다. 기존에 제안된 LSTM 기반의 Dual Encoder를 이용한 모델에서는 대화와 정답 발화에 대한 관계를 고려하지 않는 문제와 대화의 길이가 너무 길어 중간 정보의 손실되는 문제가 존재한다. 본 연구에서는 이러한 두 문제를 해결하기 위하여 ESIM구조를 통한 단어 단위의 attention, 대화의 turn별 문장 단위의 attention을 제안한다. 실험 결과 총 5000개의 검증 대화 데이터에 대하여 1 in 100 Recall@1의 성능이 37.64%로 기존 모델 대비 약 2배 높은 성능 향상을 나타내었다.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.257-262
• /
• 2009

This paper studies an automatic expansion of dialogue examples using the logs of an example-based dialogue system. Conventional approaches to example-based dialogue system manually construct dialogue examples between humans and a Chatbot, which are labor intensive and time consuming. The proposed method automatically classifies natural utterance pairs and adds them into dialogue example database. Experimental results show that lexical, POS and modality features are useful for classifying natural utterance pairs, and prove that the dialogue examples can be automatically expanded using the logs of a dialogue system.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.6
• /
• pp.263-271
• /
• 2008

This paper presents a Korean conversational agent system in a mobile environment using natural language processing techniques. The aim of a conversational agent in mobile environment is to provide natural language interface and enable more natural interaction between a human and an agent. Constructing such an agent, it is required to develop various natural language understanding components and effective utterance generation methods. To understand spoken style utterance, we perform morphosyntactic analysis, shallow semantic analysis including modality classification and predicate argument structure analysis, and to generate a system utterance, we perform example based search which considers lexical similarity, syntactic similarity and semantic similarity.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4E
• /
• pp.144-151
• /
• 2005

We evaluate the performance of emotion recognition via speech signals when a plain speaker talks to an entertainment robot. For each frame of a speech utterance, we extract the frame-based features: pitch, energy, formant, band energies, mel frequency cepstral coefficients (MFCCs), and velocity/acceleration of pitch and MFCCs. For discriminative classifiers, a fixed-length utterance-based feature vector is computed from the statistics of the frame-based features. Using a speaker-independent database, we evaluate the performance of two promising classifiers: support vector machine (SVM) and hidden Markov model (HMM). For angry/bored/happy/neutral/sad emotion classification, the SVM and HMM classifiers yield $42.3\%\;and\;40.8\%$ accuracy, respectively. We show that the accuracy is significant compared to the performance by foreign human listeners.

• MALSORI
• /
• no.56
• /
• pp.173-183
• /
• 2005

We propose a speech emotion recognition method for affective human-robot interface. In the Proposed method, emotion is classified into 6 classes: Angry, bored, happy, neutral, sad and surprised. Features for an input utterance are extracted from statistics of phonetic and prosodic information. Phonetic information includes log energy, shimmer, formant frequencies, and Teager energy; Prosodic information includes Pitch, jitter, duration, and rate of speech. Finally a pattern classifier based on Gaussian support vector machines decides the emotion class of the utterance. We record speech commands and dialogs uttered at 2m away from microphones in 5 different directions. Experimental results show that the proposed method yields $48\%$ classification accuracy while human classifiers give $71\%$ accuracy.

• Proceedings of the KSPS conference
• /
• 2005.11a
• /
• pp.163-166
• /
• 2005

We propose a speech emotion recognition method for natural human-robot interface. In the proposed method, emotion is classified into 6 classes: Angry, bored, happy, neutral, sad and surprised. Features for an input utterance are extracted from statistics of phonetic and prosodic information. Phonetic information includes log energy, shimmer, formant frequencies, and Teager energy; Prosodic information includes pitch, jitter, duration, and rate of speech. Finally a patten classifier based on Gaussian support vector machines decides the emotion class of the utterance. We record speech commands and dialogs uttered at 2m away from microphones in 5different directions. Experimental results show that the proposed method yields 59% classification accuracy while human classifiers give about 50%accuracy, which confirms that the proposed method achieves performance comparable to a human.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.6
• /
• pp.96-105
• /
• 2017

In this paper, we investigated the age classification from the speaker by analyzing the voice calls of the emergency center. We classified the adult and elderly from the call center calls using behavioral speech utterances and SVM(Support Vector Machine) which is a machine learning classifier. We selected two behavioral speech utterances through analysis of the call data from the emergency center: Silent Pause and Turn-taking latency. First, the criteria for age classification selected through analysis based on the behavioral speech utterances of the emergency call center and then it was significant(p <0.05) through statistical analysis. We analyzed 200 datasets (adult: 100, elderly: 100) by the 5 fold cross-validation using the SVM(Support Vector Machine) classifier. As a result, we achieved 70% accuracy using two behavioral speech utterances. It is higher accuracy than one behavioral speech utterance. These results can be suggested age classification as a new method which is used behavioral speech utterances and will be classified by combining acoustic information(MFCC) with new behavioral speech utterances of the real voice data in the further work. Furthermore, it will contribute to the development of the emergency situation judgment system related to the age classification.