Robust and compact multilingual word recognizers using features extracted from a phoneme similarity front-end.
Philippe MorinTed H. ApplebaumRobert BomanYi ZhaoJean-Claude JunquaPublished in: ICSLP (1998)
Keyphrases
- extracted features
- co occurrence
- extracting features
- similarity measure
- word recognition
- feature descriptors
- highly similar
- automatically extracted
- similarity function
- low level
- image features
- feature set
- pointwise mutual information
- character recognition
- information retrieval
- back end
- n gram
- binary codes
- word sense disambiguation
- speech recognition
- characteristic features
- highly discriminative
- speech recognizer
- feature space
- lexical features
- sentence similarity
- keypoints