• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.3
• /
• pp.200-207
• /
• 2001

In this paper, we propose an HMM(Hidden Markov modeJ)-MLP(Multi-layer perceptron) hybrid model for recognizing legal words on the English bank check. We adopt an explicit segmentation-based word level architecture to implement an HMM engine with nonscaled and non-normalized symbol vectors. We also introduce an MLP for implicit segmentation-based word recognition. The final recognition model consists of a hybrid combination of the HMM and MLP with a new hybrid probability measure. The main contributions of this model are a novel design of the segmentation-based variable length HMMs and an efficient method of combining two heterogeneous recognition engines. ExperimenLs have been conducted using the legal word database of CENPARMI with encouraging results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.11
• /
• pp.2800-2812
• /
• 2012

The purpose of this study was to examine the syllable-word boundary misalignment cost on word segmentation in Korean continuous speech. Previous studies have demonstrated the important role of syllabification in speech segmentation. The current study investigated whether the resyllabification process affects word recognition in Korean continuous speech. In Experiment I, under the misalignment condition, participants were presented with stimuli in which a word-final consonant became the onset of the next syllable. (e.g., /k/ in belsak ingan becomes the onset of the first syllable of ingan 'human'). In the alignment condition, they heard stimuli in which a word-final vowel was also the final segment of the syllable (e.g., /eo/ in heulmeo ingan is the end of both the syllable and word). The results showed that word recognition was faster and more accurate in the alignment condition. Experiment II aimed to confirm that the results of Experiment I were attributable to the resyllabification process, by comparing only the target words from each condition. The results of Experiment II supported the findings of Experiment I. Therefore, based on the current study, we confirmed that Korean, a syllable-timed language, has a misalignment cost of resyllabification.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.678-685
• /
• 2016

Named Entity Recognition and Classification (NERC) is a task for recognition and classification of named entities such as a person's name, location, and organization. There have been various studies carried out on Korean NERC, but they have some problems, for example lacking some features as compared with English NERC. In this paper, we propose a method that uses word embedding as features for Korean NERC. We generate a word vector using a Continuous-Bag-of-Word (CBOW) model from POS-tagged corpus, and a word cluster symbol using a K-means algorithm from a word vector. We use the word vector and word cluster symbol as word embedding features in Conditional Random Fields (CRFs). From the result of the experiment, performance improved 1.17%, 0.61% and 1.19% respectively for TV domain, Sports domain and IT domain over the baseline system. Showing better performance than other NERC systems, we demonstrate the effectiveness and efficiency of the proposed method.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.4
• /
• pp.351-359
• /
• 1991

This paper describes performances of speaker adaptation using a probabilistic spectral mapping matrix in hidden-Markov model(HMM) -based Korean isolated word recognition. Speaker adaptation based on probabilistic spectral mapping uses a well-trained prototype HMM's and is carried out by Viterbi, dynamic time warping, and forward-backward algorithms. Among these algorithms, the best performance is obtained by using the Viterbi approach together with codebook adaptation whose improvement for isolated word recognition accuracy is 42.6-68.8 %. Also, the selection of the initial values of the matrix and the normalization in computing the matrix affects the recognition accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.388-402
• /
• 2003

In this paper, we carried out the word, 4 continuous digits. continuous, and task-independent word recognition experiments to verify the effectiveness of the re-defined phoneme-likely units (PLUs) for the phonetic decision tree based HM-Net (Hidden Markov Network) context-dependent (CD) acoustic modeling in Korean appropriately. In case of the 48 PLUs, the phonemes /ㅂ/, /ㄷ/, /ㄱ/ are separated by initial sound, medial vowel, final consonant, and the consonants /ㄹ/, /ㅈ/, /ㅎ/ are also separated by initial sound, final consonant according to the position of syllable, word, and sentence, respectively. In this paper. therefore, we re-define the 39 PLUs by unifying the one phoneme in the separated initial sound, medial vowel, and final consonant of the 48 PLUs to construct the CD acoustic models effectively. Through the experimental results using the re-defined 39 PLUs, in word recognition experiments with the context-independent (CI) acoustic models, the 48 PLUs has an average of 7.06%, higher recognition accuracy than the 39 PLUs used. But in the speaker-independent word recognition experiments with the CD acoustic models, the 39 PLUs has an average of 0.61% better recognition accuracy than the 48 PLUs used. In the 4 continuous digits recognition experiments with the liaison phenomena. the 39 PLUs has also an average of 6.55% higher recognition accuracy. And then, in continuous speech recognition experiments, the 39 PLUs has an average of 15.08% better recognition accuracy than the 48 PLUs used too. Finally, though the 48, 39 PLUs have the lower recognition accuracy, the 39 PLUs has an average of 1.17% higher recognition characteristic than the 48 PLUs used in the task-independent word recognition experiments according to the unknown contextual factor. Through the above experiments, we verified the effectiveness of the re-defined 39 PLUs compared to the 48PLUs to construct the CD acoustic models in this paper.

• Phonetics and Speech Sciences
• /
• v.2 no.4
• /
• pp.153-161
• /
• 2010

Many isolated word recognition systems may generate non-similar words for recognition candidates because they use only acoustic information. The previous study [1,2] investigated several techniques which can exclude non-similar words from N-best candidate words by applying Levenstein distance measure. This paper discusses the various improving techniques of removing non-similar recognition results. The mentioned methods include comparison penalties or weights, phone accuracy based on confusion information, weights candidates by ranking order and partial comparisons. Through experimental results, it is found that some proposed method keeps more accurate recognition results than the previous method's results.

• 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.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1995.06a
• /
• pp.69-72
• /
• 1995

In this paper, a new approach for Korean connected digits recognition using the spatio-temporal neural network is reported. The data of seven digits phone numbers are used in the recognition of connected words, and in the initial experiment, digit recognition rate of 28% was achieved. In this paper, to increase recognition rate, two different approaches are analyzed. In the first system, to compensate the STNN's own defect and to emphasize the Korean word's phonic characters, the starting point of phone is pointed by comparing the average magnitude and zero-crossing rate and the ending point is pointed by comparing only zero-crossing rate. The digit recoginiton rate increased to 61%. Also, in the second system, to consider fact that same word's phone is varied severally, the number of STNN's of each word is increased from one to five, and then the varied same word's phones can be included to the increased STNN's. The digit recogniton rate of connected words increased to 89%.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.25 no.12
• /
• pp.1541-1547
• /
• 1988

In this paper, a new model for isolated word recognition called segment probability model is proposed. The proposed model is composed of two procedures of segmentation and modelling each segment. Therefore the spoken word is devided into arbitrary segments and observation probability in each segments is obtained using vector quantization. The proposed model is compared with pattern matching method and hidden Markov model by recognition experiment. The experimental results show that the proposed model is better than exsisting methods in terms of recognition rate and caculation amounts.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.5
• /
• pp.24-31
• /
• 1994

In this paper, we study continuous speech recognition of Korean language using acoustic models of word juncture coarticulation. To alleviate the performance degradation due to coarticulation problems, we use context-dependent units that model inter-word transitions in addition to intra-word transitions. In all cases the initial phone of each word has to be specified for each possible final phone of the previous word similarly for the final phone of each word. To improve the robustness of the HMM parameters, the covariance matrix is smoothed. We also use position-dependent units to improve the discriminative power between units. Simulation results show that when the improved models of word juncture coarticulation are used. the recognition performance is considerably improved compared to the baseline system using only intra-word units.