• Journal of Biomedical Engineering Research
• /
• v.29 no.4
• /
• pp.307-315
• /
• 2008

In this paper, we developed a visual-auditory cue software for reading training of motor speech disorder patients. Motor speech disorder patients can use the visual and/or auditory cues for reading training and improving their symptom. The software provides some sentences with visual-auditory cues. Our sentences used for reading training are adequately comprised on modulation training according to a professional advice in speech therapy field. To ameliorate reading skills we developed two algorithms, first one is automatically searching the starting time of speech spoken by patients and the other one is removing auditory-cue from the recorded speech that recorded at the same time. The searching of speech starting time was experimented by 10 sentences per 6 subjects in four kinds of noisy environments thus the results is that $7.042{\pm}8.99[ms]$ error was detected. The experiment of the cancellation algorithm of auditory-cue was executed from 6 subjects with 1 syllable speech. The result takes improved the speech recognition rate $25{\pm}9.547[%]$ between before and after cancellation of auditory-cue in speech. User satisfaction index of the developed program was estimated as good.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.239-246
• /
• 2023

In this paper, we propose an open-loop algorithm to classify speech and music signals using the spectral flux parameters and Mel Frequency Cepstral Coefficients(MFCC) parameters for the audio coder. To increase responsiveness, the MFCC was used as a short-term feature parameter and spectral fluxes were used as a long-term feature parameters to improve accuracy. The overall voice/music signal classification decision is made by combining the short-term classification method and the long-term classification method. The Gaussian Mixed Model (GMM) was used for pattern recognition and the optimal GMM parameters were extracted using the Expectation Maximization (EM) algorithm. The proposed long-term and short-term combined speech/music signal classification method showed an average classification error rate of 1.5% on various audio sound sources, and improved the classification error rate by 0.9% compared to the short-term single classification method and 0.6% compared to the long-term single classification method. The proposed speech/music signal classification method was able to improve the classification error rate performance by 9.1% in percussion music signals with attacks and 5.8% in voice signals compared to the Unified Speech Audio Coding (USAC) audio classification method.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.5
• /
• pp.213-221
• /
• 2006

Phone segmentation of speech waveform is especially important for concatenative text to speech synthesis which uses segmented corpora for the construction of synthetic units. because the quality of synthesized speech depends critically on the accuracy of the segmentation. In the beginning. the phone segmentation was manually performed. but it brings the huge effort and the large time delay. HMM-based approaches adopted from automatic speech recognition are most widely used for automatic segmentation in speech synthesis, providing a consistent and accurate phone labeling scheme. Even the HMM-based approach has been successful, it may locate a phone boundary at a different position than expected. In this paper. we categorized adjacent phoneme pairs and analyzed the mismatches between hand-labeled transcriptions and HMM-based labels. Then we described the dominant error patterns that must be improved for the speech synthesis. For the experiment. hand labeled standard Korean speech DB from ETRI was used as a reference DB. Time difference larger than 20ms between hand-labeled phoneme boundary and auto-aligned boundary is treated as an automatic segmentation error. Our experimental results from female speaker revealed that plosive-vowel, affricate-vowel and vowel-liquid pairs showed high accuracies, 99%, 99.5% and 99% respectively. But stop-nasal, stop-liquid and nasal-liquid pairs showed very low accuracies, 45%, 50% and 55%. And these from male speaker revealed similar tendency.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.506-511
• /
• 2018

This paper proposes an GMM(Gaussian Mixture Model)-DNN(Deep Neural Network) hybrid-based feature compensation method for effective speech recognition in noisy environments. In the proposed algorithm, the posterior probability for the conventional GMM-based feature compensation method is calculated using DNN. The experimental results using the Aurora 2.0 framework and database demonstrate that the proposed GMM-DNN hybrid-based feature compensation method shows more effective in Known and Unknown noisy environments compared to the GMM-based method. In particular, the experiments of the Unknown environments show 9.13 % of relative improvement in the average of WER (Word Error Rate) and considerable improvements in lower SNR (Signal to Noise Ratio) conditions such as 0 and 5 dB SNR.

• MALSORI
• /
• no.44
• /
• pp.105-113
• /
• 2002

In Korean, each digit is monosyllable and some pairs are known to have high confusability, causing performance degradation of connected digit recognition systems. To improve the performance, in this paper, we employ various discriminant analyses (DA) including Linear DA (LDA), Weighted Pairwise Scatter LDA WPS-LDA), Heteroscedastic Discriminant Analysis (HDA), and Maximum Likelihood Linear Transformation (MLLT). We also examine several combinations of various DA for additional performance improvement. Experimental results show that applying any DA mentioned above improves the string accuracy, but the amount of improvement of each DA method varies according to the model complexity or number of mixtures per state. Especially, more than 20% of string error reduction is achieved by applying MLLT after WPS-LDA, compared with the baseline system, when class level of DA is defined as a tied state and 1 mixture per state is used.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.15 no.9
• /
• pp.738-744
• /
• 1990

This paper proposed a real-time memory storage method and an improved sample data method from given data of the speech signal, so, the isolated word recognition system using a back-propagation learning algorithm of the neural netwrok is studied. The recognition rate and the error rate are compared with the new sample data sets generated from small sets of given sample data by the node nunber variatiion of each layer. In this result, the recognition rate of 95.1% was achived.

• ETRI Journal
• /
• v.32 no.3
• /
• pp.395-405
• /
• 2010

A novel multiclassifier system (MCS) strategy is proposed and applied to a text-dependent speaker verification task. The presented scheme optimizes the linear combination of classifiers on an on-line basis. In contrast to ordinary MCS approaches, neither a priori distributions nor pre-tuned parameters are required. The idea is to improve the most accurate classifier by making use of the incremental information provided by the second classifier. The on-line multiclassifier optimization approach is applicable to any pattern recognition problem. The proposed method needs neither a priori distributions nor pre-estimated weights, and does not make use of any consideration about training/testing matching conditions. Results with Yoho database show that the presented approach can lead to reductions in equal error rate as high as 28%, when compared with the most accurate classifier, and 11% against a standard method for the optimization of linear combination of classifiers.

• MALSORI
• /
• no.62
• /
• pp.149-161
• /
• 2007

In this paper we proposed a modified reliability function for improving bimodal speaker identification(BSI) performance. The convectional reliability function, used by N. Fox[1], is extended by introducing an optimization factor. We evaluated the proposed method in BSI domain. A BSI system was implemented based on GMM and it was tested using VidTIMIT database. Through speaker identification experiments we verified the usefulness of our proposed method. The experiments showed the improved performance, i.e., the reduction of error rate by 39%.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.5
• /
• pp.410-419
• /
• 2004

In this paper. we propose a method of confidence measure fusion under Bayesian framework for speech recognition. Centralized and distributed schemes are considered for confidence measure fusion. Centralized fusion is feature level fusion which combines the values of individual confidence scores and makes a final decision. In contrast. distributed fusion is decision level fusion which combines the individual decision makings made by each individual confidence measuring method. Optimal Bayesian fusion rules for centralized and distributed cases are presented. In isolated word Out-of-Vocabulary (OOV) rejection experiments. centralized Bayesian fusion shows over 13% relative equal error rate (EER) reduction compared with the individual confidence measure methods. In contrast. the distributed Bayesian fusion shows no significant performance increase.

• MALSORI
• /
• no.60
• /
• pp.165-180
• /
• 2006

In this paper, we present several confidence measures (CM) for speech recognition systems to evaluate the reliability of recognition results. We propose heuristic CMs such as mean log-likelihood score, N-best word log-likelihood ratio, likelihood sequence fluctuation and likelihood ratio testing(LRT)-based CMs using several types of anti-models. Furthermore, we propose new algorithms to add weighting terms on phone-level log-likelihood ratio to merge word-level log-likelihood ratios. These weighting terms are computed from the distance between acoustic models and knowledge-based phoneme classifications. LRT-based CMs show better performance than heuristic CMs excessively, and LRT-based CMs using phonetic information show that the relative reduction in equal error rate ranges between $8{\sim}13%$ compared to the baseline LRT-based CMs. We use the support vector machine to fuse several CMs and improve the performance of utterance verification. From our experiments, we know that selection of CMs with low correlation is more effective than CMs with high correlation.