ACL RD-TEC 1.0 Summarization of W04-2901
Paper Title:
A SYSTEM FOR SEARCHING AND BROWSING SPOKEN COMMUNICATIONS
A SYSTEM FOR SEARCHING AND BROWSING SPOKEN COMMUNICATIONS
Authors: Lee Begeja and Bernard Renger and Murat Saraclar and David Gibbon and Zhu Liu and Behzad Shahraray
Primarily assigned technology terms:
- algorithm
- asr system
- audio recording
- automatic speech recognition
- boundary detection
- broadcast news transcription
- broadcast news transcription system
- classification
- content analysis
- continuous speech recognition
- data mining
- database
- databases
- decision tree
- decoder
- discriminant analysis
- document retrieval
- dynamic programming
- evaluation system
- extraction method
- final state
- finite state
- finite state machine
- identification
- indexing
- information gathering
- information retrieval
- interfaces
- keyword extraction
- large vocabulary continuous speech recognition
- lattice search
- linear discriminant
- linear discriminant analysis
- lvcsr
- matching
- maximum likelihood
- mining
- mutual information estimation
- navigation
- news transcription
- parallel text alignment
- phonetic search
- porter stemming
- preprocessing
- processing
- recognition
- recognition process
- recognizer
- retrieving
- sampling
- search
- searching
- segmentation
- segmentation algorithm
- segmentation method
- sentence alignment
- skimming
- speaker identification
- speaker segmentation
- speech indexing
- speech processing
- speech recognition
- speech recognition component
- spoken document retrieval
- surveillance
- switchboard evaluation system
- tagging
- text alignment
- text search
- transcription
- user interface
- user interfaces
- visualization
- voicemail
Other assigned terms:
- acoustic models
- approach
- asr accuracy
- asr output
- backoff
- broadcast news
- broadcast news corpus
- broadcast news data
- broadcast news type
- call center
- case
- cluster
- compact representation
- composition
- context words
- continuous speech
- conversation
- corpora
- distance matrix
- distribution
- document
- document content
- document frequency
- duration
- error rate
- estimation
- evaluation test
- experimental results
- extraction process
- f-measure
- fact
- feature
- feature vector
- frame
- gaussian distribution
- gaussian mixture
- gaussian mixture models
- hypotheses
- hypothesis
- identification module
- index
- information sources
- inverse document frequency
- inverted index
- keyword
- knowledge
- language models
- lattice
- lattices
- likelihood
- linguistic
- linguistic information
- measure
- measures
- metadata
- method
- mixture models
- mutual information
- news corpus
- noise
- parallel text
- phoneme
- phrase
- pitch
- posterior
- posterior probability
- precision
- probabilities
- probability
- probability value
- process
- queries
- query
- recognition component
- recognition errors
- representations
- retrieval performance
- search results
- search space
- search term
- segments
- sentence
- sentences
- server
- substring
- switchboard corpus
- system description
- technologies
- television
- term
- term frequency
- term list
- terms
- test set
- text
- tokens
- topics
- transcript
- transcriptions
- transcripts
- tree
- trigram
- user
- user query
- video content
- vocabulary
- web page
- word
- word error rate
- word lattice
- word lattices
- word level
- word types
- words
