ACL RD-TEC 1.0 Summarization of H91-1012

Paper Title:
BYBLOS SPEECH RECOGNITION BENCHMARK RESULTS

Authors: F. Kubala and S. Austin and C. Barry and J. Makhoul and P. Placeway and R. Schwartz

Other assigned terms:

  • acoustic models
  • acoustic score
  • american english
  • approach
  • background noise
  • benchmark
  • bigram
  • bigram model
  • british english
  • byblos system
  • community
  • context models
  • corpora
  • derivation
  • detection rate
  • development set
  • dictionary
  • duration
  • entropy
  • error rate
  • evaluation data
  • evaluation set
  • evaluation test
  • events
  • fact
  • frame
  • grammar
  • grammars
  • hypotheses
  • hypothesis
  • intelligibility
  • knowledge
  • language model
  • language model score
  • language models
  • lexicon
  • mapping
  • measure
  • method
  • n-gram
  • n-gram models
  • names
  • natural language
  • nist
  • noise
  • ordered list
  • pause
  • performance evaluation
  • perplexity
  • phoneme
  • phonemes
  • probabilities
  • probability
  • procedure
  • process
  • queries
  • recognition accuracy
  • recognition error rate
  • recognition errors
  • recognition evaluation
  • rm corpus
  • sentence
  • sentences
  • signal
  • silence
  • small training corpora
  • speech corpus
  • speech data
  • speech model
  • speech recognition benchmark
  • speech recognition problem
  • spoken language
  • standard deviation
  • statistical grammar
  • statistical language model
  • statistics
  • technique
  • test data
  • test set
  • text
  • topics
  • training
  • training corpora
  • training corpus
  • training data
  • training set
  • transcriptions
  • trigram
  • triphone
  • understanding
  • utterance
  • vocabulary
  • word
  • word error rate
  • word error rates
  • word-pair grammar
  • words

Extracted Section Types:


This page last edited on 10 May 2017.

*** ***