Title:
【発明の名称】HMM装置
Document Type and Number:
Japanese Patent JP2964507
Kind Code:
B2
Abstract:
An analyzing system analyzes object signals, particularly voice signals, by estimating a generation likelihood of an observation vector sequence being a time series of feature vectors with use of a Markov model having a plurality of states and given transition probabilities from state to state. A state designation section determines a state i at a time t stochastically using the Markov model. Plural predictors, each of which is composed of a neural network and is provided per each state of the Markov model, are provided for generating a predictional vector of the feature vector xt in the state i at the time t based on values of the feature vectors other than the feature vector xt. A first calculation section calculates an error vector of the predictional vector to the feature vector xT. A second calculation section calculates a generation likelihood of the error vector using a predetermined probability distribution of the error vector according to which the error vector is generated.
Inventors:
TSUBOKA HIDEKAZU
Application Number:
JP32214889A
Publication Date:
October 18, 1999
Filing Date:
December 12, 1989
Export Citation:
Assignee:
MATSUSHITA DENKI SANGYO KK
International Classes:
G06F15/18; G06N3/00; G06N99/00; G10L15/02; G10L15/06; G10L15/08; G10L15/14; G10L15/16; G10L17/06; (IPC1-7): G10L3/00; G10L3/00
Domestic Patent References:
| JP3181999A | ||||
| JP6315299A | ||||
| JP315898A | ||||
| JP2296298A | ||||
| JP63183499A |
Other References:
【文献】米国特許5307444(US,A)
Attorney, Agent or Firm:
Fumio Iwahashi (2 others)