Improvement of noisy speech recognition using a proportional alignment decoding algorithm in the training phase

Hung, Wei-Wen; Wang, Hsiao-Chuan

doi:10.1006/csla.1998.0046

Cited by 2 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many experiments have shown that this geometrical duration distribution is not proper for the representation of the duration characteristics of the speech signal [3][4][5]. Thereby, some improvements have been made on the model by incorporating state duration with the HMM, and the variable duration HMM (VDHMM) and continuously VDH MM (CVDHMM) were proposed successively by Ferguson [6] and Levinson [7], respectively.…”

Section: Introductionmentioning

confidence: 99%

“…This kind of model is often referred to as Hidden Semi-Markov Model (HSMM; [8,9]). The commonly used duration distributions are Gamma [6,10,11], Gaussian [11], Poisson [5,8], and the uniform distribution [12], or alternatively, it can be simulated by combining some functions of the exponential function family [13]. Besides, expanded state HMM (ESHMM; [10,14]) models are also employed by some researchers to enhance the duration modeling capability of HMMs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Duration-Distribution-Based HMM for Speech Recognition

Wang

Xiao

2006

Front. Electr. Electron. Eng. China

View full text Add to dashboard Cite

To overcome the defects of the duration modeling in the homogeneous Hidden Markov Model (HMM) for speech recognition, a duration-distribution-based HMM (DDBHMM) is proposed in this paper based on a formalized definition of a left-to-right inhomogeneous Markov model. It has been demonstrated that it can be identically defined by either the state duration or the state transition probability. The speaker-independent continuous speech recognition experiments show that by only modeling the state duration in DDBHMM, a significant improvement (17.8% error rate reduction) can be achieved compared with the classical HMM. The ideal properties of DDBHMM give promise to many aspects of speech modeling, such as the modeling of the state duration, speed variation, speech discontinuity, and interframe correlation.

show abstract

Section: Introductionmentioning

confidence: 99%