• 대한음성학회지:말소리
• /
• 제52호
• /
• pp.101-110
• /
• 2004

This paper proposes a new confidence measure for utterance verification with posterior probability approximation. The proposed method approximates probabilistic likelihoods by using Viterbi search characteristics and a clustered phoneme confusion matrix. Our measure consists of the weighted linear combination of acoustic and phonetic confidence scores. The proposed algorithm shows better performance even with the reduced computational complexity than those utilizing conventional confidence measures.

• 한국음향학회지
• /
• 제21권7호
• /
• pp.632-640
• /
• 2002

본 논문에서는 기존의 RLJ-신뢰도 (RLJ-confidence measure)와 정규화 신뢰도 (normalized CM)의 단점을 보완하기 위해 ACM (Anti-filler CM)을 제안하였고, HCM (hybrid CM)을 이용하여 기존의 NCM과 제안한 ACM을 통합하였다. 제안된 ACM은 기존 신뢰도의 단점 중 하나인 오인증 (FA: false acceptance)의 원인이 반음소 모델의 구성방법에 있다고 보고 음소 인식기를 이용하여 실제 음소 수열을 추정한 다음, 이를 반음소 모델로 정의하고 신뢰도를 계산하였다. 두 가지 신뢰도의 특성을 살펴보면, 기존 NCM(FR: false rejection)에 좋은 성능을 보이고, 제안한 ACM은 FA에서 좋은 성능을 보여 두 신뢰도가 상보적인 특성을 가진다 이를 이용하여 두 가지 신뢰도를 가중치 벡터 α를 이용하여 통합하고 이를 합성 신뢰도 (HCM: Hybrid CM)라고 정의하였다. 실험결과 미검출율 (MDR: missed detection rate) 10%부근에서, HCM 적용시에 0.219 FA/KW/HR (false alarm/keyword/how)로서 NCM 단독사용에 비해 성능이 22% 향상되었다.

• 대한음성학회지:말소리
• /
• 제51호
• /
• pp.71-84
• /
• 2004

In this paper, we propose the automatic speech database verification method(or called automatic verification) based on confidence measure for a large speech database. This method verifies the consistency between given transcription and speech using the confidence measure. The automatic verification process consists of two stages : the word-level likelihood computation stage and multi-level likelihood ratio computation stage. In the word-level likelihood computation stage, we calculate the word-level likelihood using the viterbi decoding algorithm and make the segment information. In the multi-level likelihood ratio computation stage, we calculate the word-level and the phone-level likelihood ratio based on confidence measure with anti-phone model. By automatic verification, we have achieved about 61% error reduction. And also we can reduce the verification time from 1 month in manual to 1-2 days in automatic.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2006년도 추계학술대회 발표논문집
• /
• pp.3-6
• /
• 2006

This paper proposes a confidence measure employed for utterance verification in noisy environments. Most of conventional approaches estimate the proper threshold of confidence measure and apply the value to utterance rejection in recognition process. As such, their performance may degrade for noisy speech since the threshold can be changed in noisy environments. This paper presents further robust confidence measure based on the multi-pass confidence measure. The isolated word recognition based experimental results demonstrate that the proposed method outperforms conventional approaches as utterance verifier.

• 한국음향학회지
• /
• 제23권5호
• /
• pp.410-419
• /
• 2004

본 논문에서는 베이시안에 기반한 신뢰도 융합 기법을 제안한다. 음성인식에서 신뢰도는 인식 결과에 대한 신뢰의 정도를 말하며, 인식 결과가 맞는 지의 여부를 판단할 수 있다. 개별 신뢰도 기법의 신뢰도 값을 융합하여 최종 판단을 내리는 집중형 융합 방식과 개별 신뢰도 기법의 판단 결과들을 융합하는 분산형 융합의 두 가지 방식에 대해 최적의 베이시안 융합규칙이 제시되었다. 고립단어 인식에서의 미등록어 거절 실험 결과 집중형 베이시안 신뢰도 융합 기법은 개별 신뢰도 기법에 비해 13% 이상의 상대적인 에러 감소 효과를 보였으나, 분산형 베이시안 융합은 성능의 향상을 보이지 못했다.

• 한국멀티미디어학회논문지
• /
• 제19권9호
• /
• pp.1647-1658
• /
• 2016

RGB+D database has been widely used in object recognition, object tracking, robot control, to name a few. While rapid advance of active depth sensing technologies allows for the widespread of indoor RGB+D databases, there are only few outdoor RGB+D databases largely due to an inherent limitation of active depth cameras. In this paper, we propose a novel method used to build outdoor RGB+D databases. Instead of using active depth cameras such as Kinect or LIDAR, we acquire a pair of stereo image using high-resolution stereo camera and then obtain a depth map by applying stereo matching algorithm. To deal with estimation errors that inevitably exist in the depth map obtained from stereo matching methods, we develop an approach that estimates confidence of depth maps based on unsupervised learning. Unlike existing confidence estimation approaches, we explicitly consider a spatial correlation that may exist in the confidence map. Specifically, we focus on refining confidence feature with the assumption that the confidence feature and resultant confidence map are smoothly-varying in spatial domain and are highly correlated to each other. Experimental result shows that the proposed method outperforms existing confidence measure based approaches in various benchmark dataset.

• 말소리와 음성과학
• /
• 제2권3호
• /
• pp.135-139
• /
• 2010

Forensic speaker identification needs high accuracy and reliability. However, the current level of speaker identification does not reach its demand. Therefore, the confidence evaluation of results is one of the issues in forensic speaker identification. In this paper, we propose a new confidence measure of forensic speaker identification system. This is based on pitch differences between the registered utterances of the identified speaker and the test utterance. In the experiments, we evaluate this confidence measure by speech identification tasks on various environments. As the results, the proposed measure can be a good measure indicating if the result is reliable or not.

• Journal of the Korean Data and Information Science Society
• /
• 제19권4호
• /
• pp.1141-1151
• /
• 2008

Data mining is the process of sorting through large amounts of data and picking out useful information. One of the well-studied problems in data mining is the exploration of association rules. An association rule technique finds the relation among each items in massive volume database. Some interestingness measures have been developed in association rule mining. Interestingness measures are useful in that it shows the causes for pruning uninteresting rules statistically or logically. This paper propose a conditional pure confidence to evaluate association rules and then describe some properties for a proposed measure. The comparative studies with confidence and pure confidence are shown by numerical example. The results show that the conditional pure confidence is better than confidence or pure confidence.

• Journal of the Korean Data and Information Science Society
• /
• 제23권6호
• /
• pp.1127-1135
• /
• 2012

연관성 규칙 기법은 대용량데이터베이스에 있는 항목들 간의 관련성을 수치화 하는 것으로 데이터 마이닝 기법 중에서는 가장 많이 활용되고 있다. 연관성 규칙을 탐사하기 위한 연관성 규칙 평가 기준에는 지지도, 신뢰도, 향상도 등이 있다. 이들 중에서 가장 중심이 되는 신뢰도는 비대칭적 측도일 뿐만 아니라 항상 양의 값만을 취하고 있어서 항목 간에 연관성 규칙을 생성하는 데 여러가지 문제가 존재한다. 이러한 문제를 해결하기 위해 본 논문에서는 확률적 흥미도 측도 기반, 특히 주변 비율을 고려하지 않은 유사성 측도를 연관성 평가 기준으로 적용하는 방안에 대해 연구하였다. 예제에 의한 비교를 통하여 Yule과 Michael의 유사성 계수와 Pearson의 파이 계수는 신뢰도와 동일하게 연관성의 정도를 파악할 수 있는 동시에 부호를 포함하고 있어서 연관성의 방향도 알 수 있었으나, 카이 제곱 통계량 기반 측도들은 항상 양의 값만 나타날 뿐만 아니라 신뢰도와는 변화하는 양상이 다르다는 것을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권4호
• /
• pp.1824-1845
• /
• 2016

The GMM is a conventional approach which has been recently applied in many face recognition studies. However, the question about how to deal with illumination changes while ensuring high performance is still a challenge, especially with real-world databases. In this paper, we propose a Visual Observation Confidence (VOC) measure for robust face recognition for illumination changes. Our VOC value is a combined confidence value of three measurements: Flatness Measure (FM), Centrality Measure (CM), and Illumination Normality Measure (IM). While FM measures the discrimination ability of one face, IM represents the degree of illumination impact on that face. In addition, we introduce CM as a centrality measure to help FM to reduce some of the errors from unnecessary areas such as the hair, neck or background. The VOC then accompanies the feature vectors in the EM process to estimate the optimal models by modified-GMM training. In the experiments, we introduce a real-world database, called KoFace, besides applying some public databases such as the Yale and the ORL database. The KoFace database is composed of 106 face subjects under diverse illumination effects including shadows and highlights. The results show that our proposed approach gives a higher Face Recognition Rate (FRR) than the GMM baseline for indoor and outdoor datasets in the real-world KoFace database (94% and 85%, respectively) and in ORL, Yale databases (97% and 100% respectively).