ACL RD-TEC 1.0 Summarization of H92-1077
Paper Title:
SESSION 12: CONTINUOUS SPEECH RECOGNITION AND EVALUATION II
SESSION 12: CONTINUOUS SPEECH RECOGNITION AND EVALUATION II
Primarily assigned technology terms:
- acoustic modelling
- atis
- continuous speech recognition
- decoder
- fast match
- hmm csr
- language modelling
- matching
- modelling
- preprocessing
- reading
- recognition
- recognition system
- recognizer
- resource management
- sampling
- search
- speaker-independent recognition
- speech recognition
- sphinx-ii
- splitting
- stack decoder
- viterbi
- viterbi search
- wsj csr
Other assigned terms:
- benchmark
- bigram
- continuous speech
- dictionary
- error rate
- evaluation test
- language model
- language models
- lexicon
- method
- natural language
- perplexity
- procedure
- punctuation
- rapid adaptation
- running time
- search strategy
- search time
- sentences
- text
- training
- training data
- trigram
- understanding
- vocabulary
- word
- wsj corpus
