• Speech Sciences
• /
• v.15 no.2
• /
• pp.67-77
• /
• 2008

In this paper, we implemented a robust speech recognizer using the TMS320VC33 DSP. For this implementation, we had built speech and noise database suitable for the recognizer using spectral subtraction method for noise removal. The recognizer has an explicit structure in aspect that a speech signal is enhanced through spectral subtraction before endpoints detection and feature extraction. This helps make the operation of the recognizer clear and build HMM models which give minimum model-mismatch. Since the recognizer was developed for the purpose of controlling car facilities and voice dialing, it has two recognition engines, speaker independent one for controlling car facilities and speaker dependent one for voice dialing. We adopted a conventional DTW algorithm for the latter and a continuous HMM for the former. Though various off-line recognition test, we made a selection of optimal conditions of several recognition parameters for a resource-limited embedded recognizer, which led to HMM models of the three mixtures per state. The car noise added speech database is enhanced using spectral subtraction before HMM parameter estimation for reducing model-mismatch caused by nonlinear distortion from spectral subtraction. The hardware module developed includes a microcontroller for host interface which processes the protocol between the DSP and a host.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.464-470
• /
• 2016

This paper presents a speech recognition database verification system using speech measures, and describes a speech measure extraction algorithm which is applied to this system. In our previous study, to produce an effective speech quality measure for the system, we propose a combination of various speech measures which are highly correlated with WER (Word Error Rate). The new combination of various types of speech quality measures in this study is more effective to predict the speech recognition performance compared to each speech measure alone. In this paper, we increase the system independency by employing GMM acoustic score instead of HMM score which is obtained by a secondary speech recognition system. The combination with GMM score shows a slightly lower correlation with WER compared to the combination with HMM score, however it presents a higher relative improvement in correlation with WER, which is calculated compared to the correlation of each speech measure alone.

• ETRI Journal
• /
• v.23 no.4
• /
• pp.168-176
• /
• 2001

This paper discusses two important issues of corpus-based synthesis: synthesis unit generation based on phrase break strength information and pruning redundant synthesis unit instances. First, the new sentence set for recording was designed to make an efficient synthesis database, reflecting the characteristics of the Korean language. To obtain prosodic context sensitive units, we graded major prosodic phrases into 5 distinctive levels according to pause length and then discriminated intra-word triphones using the levels. Using the synthesis unit with phrase break strength information, synthetic speech was generated and evaluated subjectively. Second, a new pruning method based on weighted vector quantization (WVQ) was proposed to eliminate redundant synthesis unit instances from the synthesis database. WVQ takes the relative importance of each instance into account when clustering similar instances using vector quantization (VQ) technique. The proposed method was compared with two conventional pruning methods through objective and subjective evaluations of synthetic speech quality: one to simply limit the maximum number of instances, and the other based on normal VQ-based clustering. For the same reduction rate of instance number, the proposed method showed the best performance. The synthetic speech with reduction rate 45% had almost no perceptible degradation as compared to the synthetic speech without instance reduction.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.2
• /
• pp.37-41
• /
• 2014

The vector Taylor series (VTS) based method usually employs clean speech Hidden Markov Models (HMMs) when compensating speech feature vectors or adapting the parameters of trained HMMs. It is well-known that noisy speech HMMs trained by the Multi-condition TRaining (MTR) and the Multi-Model-based Speech Recognition framework (MMSR) method perform better than the clean speech HMM in noisy speech recognition. In this paper, we propose a method to use the noise-adapted HMMs in the VTS-based speech feature compensation method. We derived a novel mathematical relation between the train and the test noisy speech feature vector in the log-spectrum domain and the VTS is used to estimate the statistics of the test noisy speech. An iterative EM algorithm is used to estimate train noisy speech from the test noisy speech along with noise parameters. The proposed method was applied to the noise-adapted HMMs trained by the MTR and MMSR and could reduce the relative word error rate significantly in the noisy speech recognition experiments on the Aurora 2 database.

• MALSORI
• /
• no.52
• /
• pp.161-169
• /
• 2004

This paper describes the automated call routing for train ticket reservation aid system based on speech recognition. We focus on the task of automatically routing telephone calls based on user's fluently spoken response instead of touch tone menus in an interactive voice response system. Vector-based call routing algorithm is investigated and mapping table for key term is suggested. Korail database collected by KT is used for call routing experiment. We evaluate call-classification experiments for transcribed text from Korail database. In case of small training data, an average call routing error reduction rate of 14% is observed when mapping table is used.

• Phonetics and Speech Sciences
• /
• v.8 no.3
• /
• pp.31-38
• /
• 2016

Speech inversion (acoustic-to-articulatory mapping) is not a trivial problem, despite the importance, due to the highly non-linear and non-unique nature. This study aimed to investigate the performance of Deep Neural Network (DNN) compared to that of traditional Artificial Neural Network (ANN) to address the problem. The Wisconsin X-ray Microbeam Database was employed and the acoustic signal and articulatory pellet information were the input and output in the models. Results showed that the performance of ANN deteriorated as the number of hidden layers increased. In contrast, DNN showed lower and more stable RMS even up to 10 deep hidden layers, suggesting that DNN is capable of learning acoustic-articulatory inversion mapping more efficiently than ANN.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.12-20
• /
• 2003

K-SEC(Korean-Spoken English Corpus) is a kind of speech database that is being under construction by the authors of this paper. This article discusses the needs of the K-SEC from various academic disciplines and industrial circles, and it introduces the characteristics of the K-SEC design, its catalogues and contents of the recorded database, exemplifying what are being considered from both Korean and English languages' phonetics and phonologies. The K-SEC can be marked as a beginning of a parallel speech corpus, and it is suggested that a similar corpus should be enlarged for the future advancements of the experimental phonetics and the speech information technology.

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.63-71
• /
• 2020

Speech recognition is one of the areas actively commercialized using deep learning and machine learning techniques. However, the majority of speech recognition systems on the market are developed on data with limited diversity of speakers and tend to perform well on typical adult speakers only. This is because most of the speech recognition models are generally learned using a speech database obtained from adult males and females. This tends to cause problems in recognizing the speech of the elderly, children and people with dialects well. To solve these problems, it may be necessary to retain big database or to collect a data for applying a speaker adaptation. However, this paper proposes that a new end-to-end speech recognition method consists of an acoustic augmented recurrent encoder and a transformer decoder with linguistic prediction. The proposed method can bring about the reliable performance of acoustic and language models in limited data conditions. The proposed method was evaluated to recognize Korean elderly and children speech with limited amount of training data and showed the better performance compared of a conventional method.

• Proceedings of the KSPS conference
• /
• 2002.11a
• /
• pp.115-118
• /
• 2002

Neural networks have been known to have great discriminative power in pattern classification problems. In this paper, the multilayer perceptron neural networks are employed to automatically detect laryngeal pathology in speech. Also new feature parameters are introduced which can reflect the periodicity of speech and its perturbation. These parameters and cepstral coefficients are used as input of the multilayer perceptron neural networks. According to the experiment using Korean disordered speech database, incorporation of new parameters with cepstral coefficients outperforms the case with only cepstral coefficients.

• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.29-32
• /
• 2003

This paper proposes an automatic pronunciation correction system which provides users with correction guidelines for each pronunciation error. For this purpose, we develop an HMM speech recognizer which automatically classifies pronunciation errors when Korean speaks foreign language. And, we collect speech database of native and nonnative speakers using phonetically balanced word lists. We perform analysis of mispronunciation types from the experiment of automatic mispronunciation detection using speech recognizer.