• Journal of KIISE:Software and Applications
• /
• v.29 no.3
• /
• pp.167-179
• /
• 2002

In this paper, we propose a new acoustic model for characterizing segmental features and an algorithm based upon a general framework of hidden Markov models (HMMs) in order to compensate the weakness of HMM assumptions. The segmental features are represented as a trajectory of observed vector sequences by a polynomial regression function because the single frame feature cannot represent the temporal dynamics of speech signals effectively. To apply the segmental features to pattern classification, we adopted segmental HMM(SHMM) which is known as the effective method to represent the trend of speech signals. SHMM separates observation probability of the given state into extra- and intra-segmental variations that show the long-term and short-term variabilities, respectively. To consider the segmental characteristics in acoustic model, we present segmental-feature HMM(SFHMM) by modifying the SHMM. The SFHMM therefore represents the external- and internal-variation as the observation probability of the trajectory in a given state and trajectory estimation error for the given segment, respectively. We conducted several experiments on the TIMIT database to establish the effectiveness of the proposed method and the characteristics of the segmental features. From the experimental results, we conclude that the proposed method is valuable, if its number of parameters is greater than that of conventional HMM, in the flexible and informative feature representation and the performance improvement.

• MALSORI
• /
• no.43
• /
• pp.163-178
• /
• 2002

In this paper, we discuss the characteristics of Segmental-Feature HMM and summarize previous studies of SFHMM. There are several approaches to reduce the number of parameters in the previous studies. However, if the number of parameters decreased, the performance of systems also fell. Therefore, we consider the fast computation approach with preserving the same number of parameters. In this paper, we present the new segment comparison method to speed up the computation of SFHMM without loss of performance. The proposed method uses the three-frame calculation rather than the full(five) frames in the given segment. The experimental results show that the performance of the proposed system is better than that of the previous studies.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.7
• /
• pp.48-52
• /
• 2000

음성 인식에 많이 사용되는 HMM (hidden Markov model)을 개선하기 위하여 분절 특징을 사용한 분절 특징 HMM은 성능이 우수하다고 발표되었다. 그러나, 분절 길이가 증가하고 회귀 차수가 놓아질수록 분절 특징 HMM을 표현하는 매개 변수의 수도 같이 증가된다. 따라서, 본 연구에서는 상태에서 관측 가능한 분절의 분산을 분절 내의 모든 프레임에 대하여 공통적으로 표현하는 고정 분산 방법을 통하여 성능의 저하 없이 매개 변수의 수를 줄이도록 시도하였다. 실험 결과, 두 혼합 밀도인 경우 고정 분산을 이용한 분절 특징 HMM의 성능과 시변 분산을 이용한 성능의 차이가 거의 없어, 제안된 방법의 유효성을 입증하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.7
• /
• pp.641-647
• /
• 2002

In this paper, we propose the reduction method of the number of parameters in the segmental-feature HMM using trend quantization method. The proposed method shares the trend information of the polynomial trajectories by quantization. The trajectory is obtained by the sequence of feature vectors of speech signals and can be divided by trend and location information. The trend indicates the variation of consequent frame features, while the location points to the positional difference of the trajectories. Since the trend occupies the large portion of SFHMM, if the trend is shared, the number of parameters maybe decreases. To exploit the proposed system the experiments are performed on TIMIT corpus. The experimental results show that the performance of the proposed system is roughly similar to that of previous system. Therefore, the proposed system can be considered one of parameter reduction method.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.3
• /
• pp.35-44
• /
• 2000

In this paper, we propose a new trajectory model for characterizing segmental features and their interaction based upon a general framework of hidden Markov models. Each segment, a sequence of vectors, is represented by a trajectory of observed sequences. This trajectory is obtained by applying a new design matrix which includes transitional information on contiguous frames, and is characterized as a polynomial regression function. To apply the trajectory to the segmental HMM, the frame features are replaced with the trajectory of a given segment. We also propose the likelihood of a given segment and the estimation of trajectory parameters. The obervation probability of a given segment is represented as the relation between the segment likelihood and the estimation error of the trajectories. The estimation error of a trajectory is considered as the weight of the likelihood of a given segment in a state. This weight represents the probability of how well the corresponding trajectory characterize the segment. The proposed model can be regarded as a generalization of a conventional HMM and a parametric trajectory model. The experimental results are reported on the TIMIT corpus and performance is show to improve significantly over that of the conventional HMM.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.17-20
• /
• 2001

본 논문에서는 분절 특징 HMM(SFHMM)의 매개변수를 줄이는 방법을 제안한다 SFHMM이 HMM보다 우수한 성능을 보이더라도, SFHMM의 매개 변수 수는 HMM보다 많기 때문에 매개 변수 수를 줄이는 방법에 대한 연구가 필요하다. 일반적으로 궤적(trajectory)은 경향(trend) 정보와 위치(location) 정보로 분리될 수 있다. 경향은 분절 특징의 변이를 나타내며, SFHMM 변수의 많은 부분을 담당하기 때문에, 경향 정보를 공유할 수 있다면 SFHMM의 매개 변수 수는 감소될 수 있을 것이다. 제안된 방법은 궤적의 경향 정보를 양자화(quantization)에 의하여 공유한다. 제안된 방법의 성능을 살펴보기 위하여 영어 데이터베이스인 TIMIT 자료를 사용하여 실험하였다. 실험 결과 제안된 방법의 성능은 기존 연구와 거의 유사하나, 궤적의 다양한 정보를 이용한다면 궤적 정보의 공유에 의하여 매개 변수를 줄일 수 있을 것으로 보인다.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.9
• /
• pp.75-84
• /
• 2012

HMM-based Text-to-Speech systems generally utilize context dependent tri-phone units from a large corpus speech DB to enhance the synthetic speech. To downsize a large corpus speech DB, acoustically similar tri-phone units are clustered based on the decision tree using context dependent information. Context dependent information includes phoneme sequence as well as prosodic information because the naturalness of synthetic speech highly depends on the prosody such as pause, intonation pattern, and segmental duration. However, if the prosodic information was complicated, many context dependent phonemes would have no examples in the training data, and clustering would provide a smoothed feature which will generate unnatural synthetic speech. In this paper, instead of complicate prosodic information we propose a simple three prosodic boundary types and decision tree questions that use rising tone, falling tone, and monotonic tone to improve naturalness. Experimental results show that our proposed method can improve naturalness of a HMM-based Korean TTS and get high MOS in the perception test.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.24-31
• /
• 1994

In this paper, we report on the development of a speaker independent continuous speech recognition system using continuous hidden Markov models. The continuous hidden Markov model consists of mean and covariance matrices and directly models speech signal parameters, therefore does not have quantization error. Filter bank coefficients with their 1st and 2nd-order derivatives are used as feature vectors to represent the dynamic features of speech signal. We use the segmental K-means algorithm as a training algorithm and triphone as a recognition unit to alleviate performance degradation due to coarticulation problems critical in continuous speech recognition. Also, we use the one-pass search algorithm that Is advantageous in speeding-up the recognition time. Experimental results show that the system attains the recognition accuracy of $83\%$ without grammar and $94\%$ with finite state networks in speaker-indepdent speech recognition.