Proceeding of Fourth International Conference on Spoken Language Processing. ICSLP '96
DOI: 10.1109/icslp.1996.607128
|View full text |Cite
|
Sign up to set email alerts
|

Using the self-organizing map to speed up the probability density estimation for speech recognition with mixture density HMMs

Abstract: This paper presents methods to improve the probability density estimation in hidden Markov models for phoneme recognition by exploiting the Self-Organizing Map (SOM) algorithm. The advantage of using the SOM is based on the created approximative topology between the mixture densities by training the Gaussian mean vectors used as the kernel centers by the SOM algorithm. The topology makes the neighboring mixtures to respond strongly for the same inputs and so most of the nearest mixtures used to approximate the… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
6
0

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 7 publications
(6 citation statements)
references
References 10 publications
0
6
0
Order By: Relevance
“…The reason why SOM smoothing gives better initialization than other mappings (e.g. by the K-means, often with even smaller distortion) is that the generalization to the test data tends to be better and also all the mixtures contribute better to the density modelling (Kim et al, 1994;Kurimo and Somervuo, 1996).…”
Section: New Initialization Methodsmentioning
confidence: 98%
See 4 more Smart Citations
“…The reason why SOM smoothing gives better initialization than other mappings (e.g. by the K-means, often with even smaller distortion) is that the generalization to the test data tends to be better and also all the mixtures contribute better to the density modelling (Kim et al, 1994;Kurimo and Somervuo, 1996).…”
Section: New Initialization Methodsmentioning
confidence: 98%
“…A faster and better way, however, is to use the initially segmented data to train a small separate SOM for each phoneme (Kurimo and Somervuo, 1996). In this way, an equal number of mixtures and a proper representation for the variation in rare phonemes, such as /D/ and /Ö /, can be ensured.…”
Section: New Initialization Methodsmentioning
confidence: 99%
See 3 more Smart Citations