• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.1033-1038
• /
• 1994

This paper describes recognition results using the modified Learning Vector Quantization (MLVQ2) method which we proposed previously. At first, we investigated the property of duration of 29 Korean consonants and found that the variances of th duration were extremely big comparing to other languages. We carried out preliminary recognition experiments for three stop consonants P, T and K. From the recognition results, we defined the optimum conditions for the learning. Then we applied the MLVQ2 method to the recognition of Korean consonants. The training was carried out using the phoneme samples in the 611 word vocabulary uttered by 2 male speakers, where each of the speakers uttered two repetitions. The recognition experiment was carried out for the phoneme samples in two repetitions of the 611 word vocabulary uttered by another male speaker. The recognition scores for the twelve plosives were 68.2% for the test samples. The recofnition scores for the 29 Korean consonants were 64.8% for the test samples.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1991.06a
• /
• pp.58-63
• /
• 1991

This paper describes development of small scale voice dialing system using TMS320C30. Recognition vocabuliary is used 50 department name within university. In vocabulary below the middle scale, word unit recognition is more practice than phoneme unit or syllable unit recognition. In this paper, we performend recognition and model generation using DMS(Dynamic Multi-Section) and implemeted voice dialing system using TMS320C30. As a result of recognition, we achieved a 98% recognition rate in condition of section 22 and weight 0.6 and recognition time took 4 seconds.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.629-632
• /
• 2000

In this paper, we studied on the phoneme classification for Korean speech recognition. In the case of making large vocabulary speech recognition system, it is better to use phoneme than syllable or word as recognition unit. And, In order to study the difference of speech recognition according to the number of phoneme as recognition unit, we used the speech toolkit of OGI in U.S.A as recognition system. The result showed that the performance of diphthong being unified was better than that of seperated diphthongs, and we required the better result when we used the biphone than when using mono-phone as recognition unit.

• MALSORI
• /
• no.59
• /
• pp.53-68
• /
• 2006

We report the evaluation results of the Korean speech recognition platform called ECHOS. The platform has an object-oriented and reusable architecture so that researchers can easily evaluate their own algorithms. The platform has all intrinsic modules to build a large vocabulary speech recognizer: Noise reduction, end-point detection, feature extraction, hidden Markov model (HMM)-based acoustic modeling, cross-word modeling, n-gram language modeling, n-best search, word graph generation, and Korean-specific language processing. The platform supports both lexical search trees and finite-state networks. It performs word-dependent n-best search with bigram in the forward search stage, and rescores the lattice with trigram in the backward stage. In an 8000-word continuous speech recognition task, the platform with a lexical tree increases 40% of word errors but decreases 50% of recognition time compared to the HTK platform with flat lexicon. ECHOS reduces 40% of recognition errors through incorporation of cross-word modeling. With the number of Gaussian mixtures increasing to 16, it yields word accuracy comparable to the previous lexical tree-based platform, Julius.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.16 no.2
• /
• pp.251-254
• /
• 2016

An energy-efficient object matching accelerator is proposed for mobile object recognition based on matching prediction scheme. Conventionally, vocabulary tree has been used to save the external memory bandwidth in object matching process but involved massive internal memory transactions to examine each object in a database. In this paper, a novel object matching accelerator is proposed based on matching predictions to reduce unnecessary internal memory transactions by mitigating non-target object examinations, thereby improving the energy-efficiency. Experimental results show a 26% reduction in power-delay product compared to the prior art.

• MALSORI
• /
• no.45
• /
• pp.79-91
• /
• 2003

Utterance verification aims at rejecting both out-of-vocabulary (OOV) utterances and low-confidence-scored in-vocabulary (IV) utterances. For utterance verification on Korean connected digit recognition task, we investigate several methods to construct filler and anti-digit models. In particular, we propose a substitution error correction method based on 2-best decoding results. In this method, when 1st candidate is rejected, 2nd candidate is selected if it is accepted by a specific hypothesis test, instead of simply rejecting the 1st one. Experimental results show that the proposed method outperforms the conventional log likelihood ratio (LLR) test method.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.4E
• /
• pp.11-16
• /
• 1998

For large vocabulary speech recognition using HMMs, context-dependent subword units have been often employed. However, when context-dependent phone models are used, they result in a system which has too may parameters to train. The problem of too many parameters and too little training data is absolutely crucial in the design of a statistical speech recognizer. Furthermore, when building large vocabulary speech recognition systems, unseen triphone problem is unavoidable. In this paper, we propose the modified phonetic decision tree algorithm for the automatic prediction of unseen triphones which has advantages solving these problems through following two experiments in Japanese contexts. The baseline experimental results show that the modified tree based clustering algorithm is effective for clustering and reducing the number of states without any degradation in performance. The task experimental results show that our proposed algorithm also has the advantage of providing a automatic prediction of unseen triphones.

• MALSORI
• /
• no.50
• /
• pp.123-137
• /
• 2004

In this paper, we consider the implementation of connected digit recognition system using tree structured lexicon model. To implement efficiently the fixed or variable length digit recognition system, finite state network (FSN) is required. We merge the word network algorithm that implements the FSN with lexical tree search algorithm that is used for general speech recognition system for fast search and large vocabulary systems. To find the efficient modeling of digit recognition system, we investigate some performance changes when the lexical tree search is applied.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.369-372
• /
• 1998

본 논문은 어휘독립(Vocabulary-Independent) 환경에서 별도의 훈련과정 없이 인식대상 어휘를 추가 및 변경할 수 있는 가변어휘(Variable Vocabulary) 음성인식에 관한 연구를 다룬다. 가변어휘 인식은 처음에 대용량 음성 데이터베이스(DB)로 음소모델을 훈련하고 인식대상 어휘가 결정되면 발음사전에 의거하여 음소모델을 연결함으로써 별도의 훈련과정 없이 인식대상 어휘를 변경 및 추가할 수 있다. 문맥 종속형(Context-Dependent) 음소 모델인 triphone을 사용하여 인식실험을 하였고, 인식성능의 비교를 위해 어휘종속 모델을 별도로 구성하여 인식실험을 하였다. Unseen triphone 문제와 훈련 DB의 부족으로 인한 모델 파라메터의 신뢰성 저하를 방지하기 위해 state-tying 방법 중 음성학적 지식에 기반을 둔 tree-based clustering(TBC) 기법[1]을 도입하였다. Mel Frequency Cepstrum Coefficient(MFCC)와 대수에너지에 기반을 둔 3 가지 음성특징 벡터를 사용하여 인식 실험을 병행하였고, 연속 확률분포를 가지는 Hidden Markov Model(HMM) 기반의 고립단어 인식시스템을 구현하였다. 인식 실험에는 22 개 부서명 DB[3]를 사용하였다. 실험결과 어휘독립 환경에서 최고 98.4%의 인식률이 얻어졌으며, 어휘종속 환경에서의 인식률 99.7%에 근접한 성능을 보였다.

• The Magazine of the IEIE
• /
• v.28 no.4
• /
• pp.104-104
• /
• 2001