Improved open-vocabulary spoken content retrieval with word and subword lattices using acoustic feature similarity.
Hung-yi LeePo-wei ChouLin-Shan LeePublished in: Comput. Speech Lang. (2014)
Keyphrases
- out of vocabulary
- spoken document retrieval
- n gram
- language model
- word segmentation
- vocabulary tree
- similarity estimation
- similarity matching
- text queries
- broadcast news
- pointwise mutual information
- similarity assessment
- similarity retrieval
- keywords
- cross language information retrieval
- named entity recognition
- retrieval method
- test collection
- index terms
- retrieval model
- query terms
- prosodic features
- image similarity
- term frequency
- information retrieval
- similarity measure
- co occurrence
- similarity measurement
- speech recognition systems
- information retrieval systems
- document similarity
- speech recognition
- parallel corpora
- sentence similarity
- distance measure
- text classification
- query expansion
- cross language
- partial matching
- document retrieval
- text retrieval
- visual similarity
- word recognition
- term weighting
- cross lingual
- bag of words
- named entities
- document space
- shape similarity
- word similarity
- edit distance
- information extraction
- feature vectors