• The Journal of the Acoustical Society of Korea
• /
• v.20 no.6
• /
• pp.46-50
• /
• 2001

This paper proposes a new recognition time reduction algorithm Score-Cache technique, which is applicable to the HMM-base speech recognition system. Score-Cache is a very unique technique that has no other performance degradation and still reduces a lot of search time. Other search reduction techniques have trade-offs with the recognition rate. This technique can be applied to the continuous speech recognition system as well as the isolated word speech recognition system. W9 can get high degree of recognition time reduction by only replacing the score calculating function, not changing my architecture of the system. This technique also can be used with other recognition time reduction algorithms which give more time reduction. We could get 54% of time reduction at best.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.6
• /
• pp.53-61
• /
• 2000

In the speech recognition under the noisy environments, reducing the mismatch introduced between training and testing environments is an important issue. Spectral subtraction is widely used technique because of its simplicity and relatively good performance in noisy environments. In this paper, we introduce histogram method as a reliable noise estimation approach for spectral subtraction. This method has advantages over the conventional noise estimation methods in that it does not need to detect non-speech intervals and it can estimate the noise spectra even in time-varying noise environments. Even though spectral subtraction is performed using a reliable average noise spectrum by the histogram method, considerable amount of residual noise remains due to the variations of instantaneous noise spectrum about mean. To overcome this limitation, we propose a new over-estimation technique based on distribution characteristics of histogram used for noise estimation. Since the proposed technique decides the degree of over-estimation adaptively according to the measured noise distribution, it has advantages to be few the influence of the SNR variation on the noise levels. According to speaker-independent isolated word recognition experiments in car noise environment under various SNR conditions, the proposed histogram-based over-estimation technique outperforms the conventional over-estimation technique.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.1
• /
• pp.28-33
• /
• 2009

In the research of speech recognition, locating the beginning and end of a speech utterance in a background of noise is of great importance. The conventional methods for speech endpoint detection are based on two simple time-domain measurements-short-time energy, and short-time zero-crossing rate, which couldn't guarantee the precise results if in the low signal-to-noise ratio environments. This paper proposes a novel approach that finds the Lyapunov exponent of time-domain waveform. This proposed method has no use for obtaining the frequency-domain parameters for endpoint detection process, e.g. Mel-Scale Features, which have been introduced in other paper. Accordingly, this algorithm is low complexity and suitable for Digital Isolated Word Recognition System.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.5
• /
• pp.5-10
• /
• 1995

In this paper we present a clustering procedure based on the use of HMM in order to get multiple statistical models which can well absorb the variants of each speaker with different ways of saying words. The HMM-clustered models obtained from the developed technique are applied to the speaker independent isolated word recognition. The HMM clustering method splits off all observation sequences with poor likelihood scores which fall below threshold from the training set and create a new model out of the observation sequences in the new cluster. Clustering is iterated by classifying each observation sequence as belonging to the cluster whose model has the maximum likelihood score. If any clutter has changed from the previous iteration the model in that cluster is reestimated by using the Baum-Welch reestimation procedure. Therefore, this method is more efficient than the conventional template-based clustering technique due to the integration capability of the clustering procedure and the parameter estimation. Experimental data show that the HMM-based clustering procedure leads to $1.43\%$ performance improvements over the conventional template-based clustering method and $2.08\%$ improvements over the single HMM method for the case of recognition of the isolated korean digits.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.2
• /
• pp.90-97
• /
• 2000

This paper describes an speaker-adaptive speech recognition system which make a reliable recognition of speech signal for new speakers. In the Proposed method, an speech spectrum of new speaker is adapted to the reference speech spectrum by using Parameters of a 1st linear transformation network at the front of phoneme classification neural network. And the recognition system is based on semicontinuous HMM(hidden markov model) which use the multilayer perceptron as a fuzzy vector quantizer. The experiments on the isolated word recognition are performed to show the recognition rate of the recognition system. In the case of speaker adaptation recognition, the recognition rate show significant improvement for the unadapted recognition system.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.56-62
• /
• 2001

In this paper, we adopt a missing data theory to speech recognition. It can be used in order to maintain high performance of speech recognizer when the missing data occurs. In general, hidden Markov model (HMM) is used as a stochastic classifier for speech recognition task. Acoustic events are represented by continuous probability density function in continuous density HMM(CDHMM). The missing data theory has an advantage that can be easily applicable to this CDHMM. A marginalization method is used for processing missing data because it has small complexity and is easy to apply to automatic speech recognition (ASR). Also, a spectral subtraction is used for detecting missing data. If the difference between the energy of speech and that of background noise is below given threshold value, we determine that missing has occurred. We propose a new method that examines the reliability of detected missing data using voicing probability. The voicing probability is used to find voiced frames. It is used to process the missing data in voiced region that has more redundant information than consonants. The experimental results showed that our method improves performance than baseline system that uses spectral subtraction method only. In 452 words isolated word recognition experiment, the proposed method using the voicing probability reduced the average word error rate by 12％ in a typical noise situation.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.2
• /
• pp.96-100
• /
• 2006

Kim et al. Proposed Normalized Confidence Measure (NCM) [1-2] and it was successfully used for rejecting mis-recognized words in isolated word recognition. However their experiments were performed on the fixed word speech recognition. In this Paper we apply NCM to the domain of vocabulary independent speech recognition (VISP) and shows the rejection Performance of NCM in VISP. Specialty we Propose vector quantization (VQ) based method for overcoming the problem of unseen triphones. It is because NCM uses the statistics of triphone confidence in the case of triphone-based normalization. According to speech recognition experiments Phone-based normalization method shows better results than RLJC[3] and also triphone-based normalization approach. This results are different with those of Kim et al [1-2]. Concludingly the Phone-based normalization shows robust Performance in VISP domain.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.5
• /
• pp.8-14
• /
• 2009

In the research field of speech recognition, pinpointing the endpoints of speech utterance even with the presence of background noise is of great importance. These noise present during recording introduce disturbances which complicates matters since what we just want is to get the stationary parameters corresponding to each speech section. One major cause of error in automatic recognition of isolated words is the inaccurate detection of the beginning and end boundaries of the test and reference templates, thus the necessity to find an effective method in removing the unnecessary regions of a speech signal. The conventional methods for speech endpoint detection are based on two linear time-domain measurements: the short-time energy, and short-time zero-crossing rate. They perform well for clean speech but their precision is not guaranteed if there is noise present, since the high energy and zero-crossing rate of the noise is mistaken as a part of the speech uttered. This paper proposes a novel approach in finding an apparent threshold between noise and speech based on Lyapunov Exponents (LEs). This proposed method adopts the nonlinear features to analyze the chaos characteristics of the speech signal instead of depending on the unreliable factor-energy. The excellent performance of this approach compared with the conventional methods lies in the fact that it detects the endpoints as a nonlinearity of speech signal, which we believe is an important characteristic and has been neglected by the conventional methods. The proposed method extracts the features based only on the time-domain waveform of the speech signal illustrating its low complexity. Simulations done showed the effective performance of the Proposed method in a noisy environment with an average recognition rate of up 92.85% for unspecified person.

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.42 no.1
• /
• pp.13-20
• /
• 2005

In this paper, I recognized Korean isolated words using CVEM which is suggested by M. Bilginer et al. CVEM is an algorithm which is easy to extract the common properties from training voice signals and also doesn't need complex calculation. In addition CVEM shows high accuracy in recognition results. But, CVEM has couple of problems which are impossible to use for many training voices and no discriminant information among extracted common vectors. To get the optimal common vectors from certain voice classes, various voices should be used for training. But CVEM is impossible to get continuous high accuracy in recognition because CVEM has a limitation to use many training voices and the absence of discriminant information among common vectors can be the source of critical errors. To solve above problems and improve recognition rate, k-clustering subspace method and DCVEM suggested. And did various experiments using voice signal database made by ETRI to prove the validity of suggested methods. The result of experiments shows improvements in performance. And with proposed methods, all the CVEM problems can be solved with out calculation problem.

• Speech Sciences
• /
• v.15 no.4
• /
• pp.85-96
• /
• 2008

Automatic speech recognition has severe performance degradation under noisy environments. To cope with the noise problem, many methods have been proposed. Most of them focused on noise-robust features or model adaptation. However, researchers have overlooked utterance verification (UV) under noisy environments. In this paper we discuss UV problems based on the normalized confidence measure. First, we show that UV performance is also degraded in noisy environments with the experiments of an isolated word recognition. Then we observe how the degradation of UV performances is suffered. Based on the UV experiments we propose a modeling method of the statistics of phone confidences using sigmoid functions. For obtaining the parameters of the sigmoidal models, the particle swarm optimization (PSO) is adopted. The proposed method improves 20% rejection performance. Our experimental results show that the PSO-NCM can apply noise speech recognition successfully.