ACL RD-TEC 1.0 Summarization of H93-1022
Paper Title:
IMPROVED KEYWORD-SPOTTING USING SRI'S DECIPHER(TM) LARGE-VOCABUARLY SPEECH-RECOGNITION SYSTEM
IMPROVED KEYWORD-SPOTTING USING SRI'S DECIPHER(TM) LARGE-VOCABUARLY SPEECH-RECOGNITION SYSTEM
Primarily assigned technology terms:
- acoustic model training
- acoustic modeling
- algorithm
- atis
- beam search
- bigram estimation
- computing
- continuous speech recognition
- continuous-speech recognition
- database
- decoding
- estimation algorithm
- finite state
- human-computer interaction
- hypothesizing
- information system
- katz back-off bigram estimation
- keyword spotting
- keyword-spotting \
- language model training
- language modeling
- language training
- large-vocabulary continuous-speech recognition
- large-vocabulary recognition
- model estimation
- model training
- modeling
- normalization
- recognition
- recognition system
- recognition system \
- recognition systems
- recognizer
- scoring
- search
- searching
- smoothing
- speaker-independent speech recognition
- speech recognition
- speech recognition system
- speech recognizer
- speech-recognition
- time-synchronous beam search
- transcription
- viterbi
- weighting
- word-spotting
- word-spotting task
- wordspotting
Other assigned terms:
- acoustic model
- acoustic models
- approach
- background model
- beam
- bigram
- bigram language model
- cepstral energy
- content words
- continuous speech
- continuous-speech
- conversation
- detection rate
- duration
- estimation
- experimental results
- finite state grammar
- grammar
- hypotheses
- hypothesis
- information need
- keyword
- language model
- lexicon
- likelihood
- log-likelihood
- meaning
- method
- perplexity
- probabilities
- probability
- queries
- query
- recognition task
- signal
- speech corpus
- speech signal
- spoken language
- spoken language applications
- spontaneous conversation
- switchboard corpus
- technology
- test set
- text
- tm system
- topics
- topology
- training
- training data
- training set
- transcriptions
- transition probabilities
- triphone
- understanding
- utterance
- viterbi path
- vocabulary
- vocabulary size
- word
- word model
- word sequences
- words
