ACL RD-TEC 1.0 Summarization of A00-1044
Paper Title:
NAMED ENTITY EXTRACTION FROM NOISY INPUT: SPEECH AND OCR
NAMED ENTITY EXTRACTION FROM NOISY INPUT: SPEECH AND OCR
Authors: David Miller and Sean Boisen and Richard Schwartz and Rebecca Stone and Ralph Weischedel
Primarily assigned technology terms:
- acoustic modelling
- algorithm
- asr system
- automatic speech recognition
- capitalization
- character recognition
- computing
- database
- entity extraction
- extraction system
- extraction systems
- extraction technology
- finite state
- hidden markov
- hidden markov model
- information extraction
- information extraction system
- language model training
- language modeling
- markov model
- message understanding
- model training
- modeling
- modelling
- name finding
- named entity extraction
- ne extraction
- ne system
- optical character recognition
- orthographic representation
- recognition
- recognizer
- rule-based system
- search
- search algorithm
- speech recognition
- speech recognizer
- speech systems
- transcription
Other assigned terms:
- abbreviations
- annotation
- approach
- asr transcript
- bigram
- bigram language model
- broadcast news
- broadcast news data
- case
- case information
- characters
- data set
- document
- english text
- error rate
- evaluations
- events
- f measure
- f-measure
- fact
- implementation
- input text
- knowledge
- language model
- large vocabulary speech
- lexicon
- measure
- message
- message understanding conference
- named entity
- names
- ne task
- nist
- noisy input
- perplexity
- probability
- proper names
- punctuation
- punctuation marks
- sentence
- set size
- speech input
- speech model
- technology
- test data
- test set
- text
- training
- training data
- training set
- training set size
- training size
- transcribed speech
- transcript
- transcriptions
- transcripts
- understanding
- vocabulary
- vocabulary size
- word
- word error rate
- word error rates
- word level
- words
