ACL RD-TEC 1.0 Summarization of P04-1086
Paper Title:
USING CONDITIONAL RANDOM FIELDS TO PREDICT PITCH ACCENTS IN CONVERSATIONAL SPEECH
USING CONDITIONAL RANDOM FIELDS TO PREDICT PITCH ACCENTS IN CONVERSATIONAL SPEECH
Authors: Michelle Gregory and Yasemin Altun
Primarily assigned technology terms:
- accent prediction
- algorithm
- automatic speech recognition
- classification
- coding
- conditional random fields
- crfs
- crossvalidation
- database
- decoding
- feature extraction
- feature representation
- hmms
- identification
- iterative scaling
- logistic regression
- loss function
- markov model
- parameter estimation
- pitch accent placement
- pitch accent prediction
- recognition
- regression
- sequence labeling
- speech recognition
- speech synthesis
- synthesis
- textual understanding
- transcription
- viterbi
- viterbi algorithm
- viterbi decoding
Other assigned terms:
- accent
- accent placement
- acoustic information
- acoustic models
- american english
- bigram
- break
- case
- citation
- coding scheme
- collocation
- conditional probabilities
- conditional probability
- conditional probability distribution
- contextual information
- conversational speech
- corpora
- correlation
- correlations
- crf model
- distribution
- duration
- estimation
- events
- fact
- feature
- function words
- gaussian prior
- hmm model
- implementation
- information content
- intonational phrase
- joint probability
- kappa
- knowledge
- labeling
- likelihood
- measure
- measures
- method
- mutual information
- norm
- open domain
- part of speech
- part-of-speech
- part-of-speech tag
- pause
- pauses
- phrase
- pitch
- pitch accent
- prediction model
- prediction task
- probabilities
- probability
- probability distribution
- pronunciation
- rank correlation
- representations
- sentence
- sentences
- sequence model
- silence
- sources of information
- switchboard corpus
- syllables
- syntactic category
- syntactic features
- target word
- term
- terms
- text
- textual information
- textual input
- topics
- training
- training set
- understanding
- unigram
- utterance
- verb
- window size
- word
- word frequency
- word level
- words
