A proposal for an evaluation set selection method for corpus-based speech translation.
Fumiaki SugayaKeiji YasudaSeiichi YamamotoPublished in: Systems and Computers in Japan (2005)
Keyphrases
- method finds
- detection method
- preprocessing
- computational complexity
- classification accuracy
- significant improvement
- high accuracy
- evaluation model
- support vector machine svm
- small number
- dynamic programming
- objective function
- similarity measure
- selection scheme
- multiscale
- hybrid method
- gold standard
- evaluation method
- high precision
- segmentation method
- synthetic data
- clustering method
- pairwise
- markov random field
- feature set
- neural network
- feature vectors
- support vector machine
- prior knowledge