Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

HMM2- A Novel Approach to HMM Emission Probability Estimation

Katrin Weber (1,2), Samy Bengio (1), Hervé Bourlard (1,2)

(1) IDIAP - Dalle Molle Institute of Perceptual Artificial Intelligence, Martigny, Switzerland
(2) EPFL - Swiss Federal Institute of Technology, Lausanne, Switzerland

In this paper, we discuss and investigate a new method to estimate local emission probabilities in the framework of hidden Markov models (HMM). Each feature vector is considered to be a sequence and is supposed to be modeled by yet another HMM. Therefore, we call this approach ‘HMM2’. There is a variety of possible topologies of such HMM2 systems, e.g. incorporating trellis or ergodic HMM structures. Preliminary HMM2 speech recognition experiments on cepstral and spectral features yielded worse results than state-of-the-art systems. However, we believe that HMM2 systems have a lot of potential advantages and are therefore worth investigating further.

Full Paper

Bibliographic reference.  Weber, Katrin / Bengio, Samy / Bourlard, Hervé (2000): "HMM2- a novel approach to HMM emission probability estimation", In ICSLP-2000, vol.3, 147-150.