ACL RD-TEC 1.0 Summarization of W97-1008
Paper Title:
WHAT MAKES A WORD: LEARNING BASE UNITS IN JAPANESE FOR SPEECH RECOGNITION
WHAT MAKES A WORD: LEARNING BASE UNITS IN JAPANESE FOR SPEECH RECOGNITION
Authors: Laura Mayfield Tomokiyo and Klaus Ries
Primarily assigned technology terms:
- algorithm
- approximation
- automated process
- chunker
- chunking
- clustering
- computational natural language learning
- database
- databases
- decoder
- dictionary modification
- grouping
- japanese speech recognition
- language learning
- language modeling
- learning
- model estimation
- modeling
- morphological analysis
- natural language learning
- parsing
- pattern recognition
- phrase finding
- predictor
- processing
- reading
- recognition
- recognition system
- recognition systems
- recognizer
- romanization
- search
- segmentation
- segmentation algorithm
- speech recognition
- speech recognition system
- speech recognition systems
- speech recognizer
- speech system
- statistical language modeling
- tagging
- tokenization
- tokenizer
- translation system
- viterbi
- word grouping
Other assigned terms:
- acoustic model
- acoustic signal
- alphabet
- appointment scheduling
- approach
- auxiliary verb
- auxiliary verbs
- backoff
- bias
- bigram
- bigram model
- break
- bunsetsu
- casual speech
- characters
- chunk
- chunks
- clusters
- community
- composition
- compounding
- compounds
- corpora
- dialogues
- dictionary
- dictionary entry
- distribution
- document
- entropy
- estimation
- evaluations
- fact
- formal speech
- generation
- generation process
- grammar
- implementation
- inflected forms
- inflection
- input string
- japanese language
- japanese text
- kanji
- knowledge
- language model
- language model perplexity
- language modeling toolkit
- language models
- manual segmentation
- mapping
- meaning
- measure
- measures
- method
- model perplexity
- modeling power
- modeling problem
- modeling toolkit
- morphemes
- mutual information
- natural language
- natural speech
- noise
- noun compounding
- nouns
- orthography
- perplexity
- phoneme
- phonemes
- phrase
- predictive power
- probability
- procedure
- process
- pronunciation
- pronunciation dictionary
- recognition accuracy
- representations
- search problem
- search space
- segments
- semantic
- sentence
- sentences
- signal
- slot
- speech recognition problem
- spoken language
- spontaneous scheduling task
- statistical language model
- statistical model
- statistical models
- stem
- stems
- style
- suffix
- syllables
- symbol
- symbols
- tagging scheme
- technique
- terms
- test corpora
- test corpus
- test set
- text
- theorem
- toolkit
- training
- training corpus
- transcriptions
- trigram
- verb
- verb meaning
- verb stem
- vocabulary
- vocabulary growth
- vocabulary size
- vowel
- word
- word boundaries
- word choice
- word classes
- word dictionary
- word sequence
- word sequences
- word types
- words
