Combination of phone N-grams for a MPEG-7-based spoken document retrieval system.
Nicolas MoreauHyoung-Gook KimThomas SikoraPublished in: EUSIPCO (2004)
Keyphrases
- n gram
- retrieval systems
- information retrieval systems
- language model
- information retrieval
- web documents
- text classification
- bag of words
- variable length
- text documents
- language independent
- multimedia
- word level
- speech recognition
- document representation
- relevance ranking
- language modelling
- document images
- language modeling
- test collection
- word segmentation
- document ranking
- relevance feedback
- databases
- document collections
- inside outside algorithm
- automatic speech recognition
- part of speech
- information access
- ranked list
- retrieval model
- relevant documents
- text mining