• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.167-170
• /
• 2000

일반적인 한국어 대용량 어휘인식에 사용되는 triphone 모델은 한국어의 특성을 잘 표현한다는 장점이 있으나 인식시간이 길어지게 된다. 이러한 triphone 모델의 단점을 극복하기 위해 음절단위 HMM 모델을 사용하는 방법이 있는데 이 모델은 인식시간을 줄일 수 있으나 triphone 모델에 비해서 인식률이 낮다. 본 논문에서는 음성 인식시간을 단축시키고 조음현상을 고려하기 위하여 초성과 종성 자음은 각각의 biphones으로 나타내고 중성 모음은 1개의 monophone으로 나타내는 모델을 제안하였다. PBW445 음성 데이터베이스에 대한 실험결과, 제안한 인식모델이 triphone 모델에 가까운 인식률을 나타내었으며, 인식시간을 크게 단축하였다.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.37-38
• /
• 2007

This paper reviews the issues in implementing sound recognizers in real environments. First is the signal corruption caused by background noises and reverberation. Second is the open-set problem which is the problem of rejecting out-of-vocabulary words and noises. These two issues must be solved for noise robust recognizers.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1995.06a
• /
• pp.65-68
• /
• 1995

이 논문은 현재 주식시장에 상장되어 있는 약 700개 회사의 현재주가를 음성인식을 이용하여 검색할 수 있는 대어휘, 화자독립, 단어독립 음성인식 시스팀의 운용자를 위한 운용관리 시스팀에 대해 기술하였다. KT-STOCK은 시스팀의 음성안내에 따라 사용자가 전화기에 상장회사 이름을 말하면, 이 시스팀은 그 회사의 현재 증권정보를 말해준다. 이 시스팀의 운용관리 시스팀은 주식시장에 상장된 종목의 변화에 따라서 인식대상 단어를 추가하거나 삭제, 조회할 때 그 처리를 용이하게 할 수 있도록 구현되었다.

• 중국학논총
• /
• no.66
• /
• pp.45-61
• /
• 2020

Chinese for Journalism and Communication (CJC) is used not only to communicate, gather, read and understand the news and information in the daily life, but also to work as a professional in journalism. It includes daily CJC and professional CJC. It has similarities and difference with journalism terms. The lexical values of CJC words includes: recognition value, frequency value, composition value, and aggregation value.

• Journal of the Korea Convergence Society
• /
• v.7 no.6
• /
• pp.7-11
• /
• 2016

Voice recognition systems which use a white noise and voice recognition environment are not correct voice recognition with variable voice mixture. Therefore in this paper, we propose a method using the convergence of Bayesian technique and selecting voice for effective voice recognition. we make use of bank frequency response coefficient for selective voice extraction, Using variables observed for the combination of all the possible two observations for this purpose, and has an voice signal noise information to the speech characteristic extraction selectively is obtained by the energy ratio on the output. It provide a noise elimination and recognition rates are improved with combine voice recognition of bayesian methode. The result which we confirmed that the recognition rate of 2.3% is higher than HMM and CHMM methods in vocabulary recognition, respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.2
• /
• pp.196-205
• /
• 1991

This paper is a study on the isolated word recognition of speaker independent which proposes to newly improved MSVQ(multisection vector quantization) recognition system which improve the classical MSVQ recognition system. It is a difference that test pattern has on more section than reference pattern in recognition system 146 DDD area names are selected as recognition vocabulary. 12th LPC cepstral coefficients is used as feature parameter. and when codebook is generated, MINSUM and MINMAX are used in finding the centroid. According to the experiment result. it is proved that this method is better than VQ(vector quantization) recognition methods, DTW(dynamic time warping) pattern matching methods and classical MSVQ methods for recognition rate and recognition time.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.6
• /
• pp.47-52
• /
• 2000

In the case of making large vocabulary speech recognition system, it is better to use the segment than the syllable or the word as the recognition mit. In this paper, we study on the proper recognition units for Korean speech recognition. For experiments, we use the speech toolkit of OGI in U.S.A. The result shows that the recognition rate of the case in which the diphthong is established as a single unit is superior to that of the case in which the diphthong is established as two units, i.e. a glide plus a vowel. And also, the recognition rate of the case in which the biphone is used as the recognition unit is better than that of the case in which the mono-phoneme is used.

• MALSORI
• /
• no.59
• /
• pp.27-51
• /
• 2006

For speech recognition systems, rejection function as well as decoding function is necessary to improve the reliability. There have been many research efforts on out-of-vocabulary word rejection, however, little attention has been paid on non-target sentence rejection. Recently pronunciation approaches using speech recognition increase the need for non-target sentence rejection to provide more accurate and robust results. In this paper, we proposed filler model method and word/phoneme detection ratio method to implement non-target sentence rejection system. We made performance evaluation of filler model along to word-level, phoneme-level, and sentence-level filler models respectively. We also perform the similar experiment using word-level and phoneme-level word/phoneme detection ratio method. For the performance evaluation, the minimized average of FAR and FRR is used for comparing the effectiveness of each method along with the number of words of given sentences. From the experimental results, we got to know that word-level method outperforms the other methods, and word-level filler mode shows slightly better results than that of word detection ratio method.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1291-1295
• /
• 2004

This paper proposed a new approach for endpoint detection of isolated speech, which proves to significantly improve the endpoint detection performance. The proposed algorithm relies on the root mean square energy (rms energy), zero crossing rate and spectral characteristics of the speech signal where the Euclidean distance measure is adopted using cepstral coefficients to accurately detect the endpoint of isolated speech. The algorithm offers better performance than traditional energy-based algorithm. The vocabulary for the experiment includes English digit from one to nine. These experimental results were conducted by 360 utterances from a male speaker. Experimental results show that the accuracy of the algorithm is quite acceptable. Moreover, the computation overload of this algorithm is low since the cepstral coefficients parameters will be used in feature extraction later of speech recognition procedure.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.83-86
• /
• 1998

This paper describes an overview of the system designed for 500-word speech recognition. The system is based on the triphone models and uses Dynamic Multisection(DMS) technique for pattern matching. The system is very flexible in the sense of the word-dictionary which is changable spontaneously without any training. The vocabulary selected for the experiments is 561 words of province names, district names of Seoul and Pusan. The experimental results which will be shown here are preliminary because only one speaker was involved in the experiments. But the result is satisfactory when we see the performance is 95.1%. The system is designed on the Windows-95 and works in realtime on the Pentium-133 computer.