ACL RD-TEC 1.0 Summarization of N06-2023
Paper Title:
SUMMARIZING SPEECH WITHOUT TEXT USING HIDDEN MARKOV MODELS
SUMMARIZING SPEECH WITHOUT TEXT USING HIDDEN MARKOV MODELS
Authors: Sameer Maskey and Julia Hirschberg
Primarily assigned technology terms:
- algorithm
- asr transcription
- automatic speech recognition
- classifiers
- computing
- decoding
- decomposition
- exact matching
- extractive summarization
- hidden markov
- hidden markov model
- hidden markov models
- hmms
- information retrieval
- language processing
- markov model
- matching
- maximum likelihood
- modeling
- objective evaluation
- processing
- recognition
- segmentation
- sentence detection
- speaker segmentation
- speech recognition
- speech summarization
- summarization
- summarization systems
- summarizer
- text summarization
- training procedure
- transcription
- viterbi
- viterbi algorithm
Other assigned terms:
- anchors
- annotator
- approach
- broadcast news
- distribution
- document
- document text
- duration
- evaluation metrics
- f-measure
- feature
- geometric distribution
- gold standard
- hmm model
- human performance
- independence assumption
- information retrieval measure
- language processing tasks
- lattice
- lexical features
- lexical information
- likelihood
- markov models
- meaning
- measure
- method
- pause
- pitch
- posterior
- posterior probability
- precision
- probabilities
- probability
- procedure
- processing tasks
- prosody
- relative frequency
- segments
- sentence
- sentence boundaries
- sentence position
- sentences
- signal
- speaking rate
- speech prosody
- test set
- text
- topic shift
- topics
- topology
- training
- transcript
- transcripts
- transition probabilities
- word
- words
