ACL RD-TEC 1.0 Summarization of W04-2907
Paper Title:
GENERAL INDEXATION OF WEIGHTED AUTOMATA - APPLICATION TO SPOKEN UTTERANCE RETRIEVAL
GENERAL INDEXATION OF WEIGHTED AUTOMATA - APPLICATION TO SPOKEN UTTERANCE RETRIEVAL
Authors: Cyril Allauzen and Mehryar Mohri and Murat Saraclar
Primarily assigned technology terms:
- algorithm
- approximation
- automatic speech recognizer
- automaton
- classification
- decoder
- evaluation system
- filter transducer
- final state
- finite automata
- finite-state transducer
- full indexing
- index construction
- indexation algorithm
- indexation algorithm and framework
- indexing
- language processing
- language processing system
- matching
- natural language processing
- natural language processing system
- optimization
- preprocessing
- probability semiring
- processing
- pruning
- reading
- recognition
- recognizer
- regular expression
- retrieving
- search
- searching
- speech indexation
- speech indexing
- speech processing
- speech recognition
- speech recognizer
- string matching
- switchboard evaluation system
- text indexation
- transducer
- transducers
- transduction
- viterbi
- weight pushing
- weighted automata
- weighted determinization
- weighted finite-state transducer
- weighted transducer
Other assigned terms:
- alphabet
- approach
- automata
- broadcast news
- broadcast news corpus
- case
- composition
- conversation
- corpora
- dictionary
- distribution
- edit distance
- evaluation metrics
- evaluation test
- experimental results
- f-measure
- finite set
- grapheme
- hypotheses
- implementation
- index
- information sources
- input string
- labeling
- lattice
- lattices
- linear time
- log-likelihood
- mapping
- maps
- method
- named entities
- natural language
- negation
- news corpus
- phoneme
- posterior
- precision
- probabilities
- probability
- probability distribution
- procedure
- process
- projection
- pronunciation
- pronunciation dictionary
- pruning threshold
- queries
- query
- regular expressions
- representations
- retrieval performance
- search procedure
- search results
- segments
- semiring
- statistics
- substring
- suffix
- suffixes
- switchboard corpus
- syntactic information
- technique
- test set
- text
- tokens
- topics
- transcript
- transcriptions
- tropical semiring
- user
- utterance
- vocabulary
- word
- word lattice
- word lattices
- word sequences
- word strings
- word types
- words
