ACL RD-TEC 1.0 Summarization of W04-2902
Paper Title:
ANALYSIS AND PROCESSING OF LECTURE AUDIO DATA: PRELIMINARY INVESTIGATIONS
ANALYSIS AND PROCESSING OF LECTURE AUDIO DATA: PRELIMINARY INVESTIGATIONS
Authors: James Glass and Timothy J. Hazen and Lee Hetherington and Chao Wang
Primarily assigned technology terms:
- audio indexing
- computer science
- continuous speech recognition
- corpus creation
- data collection
- educational technology
- human language
- human language technology
- indexing
- language model training
- language modeling
- language modelling
- language processing
- language technology
- large-vocabulary continuous speech recognition
- large-vocabulary continuous speech recognition technology
- lecture processing
- matching
- matrix inversion
- matrix multiplication
- model training
- modeling
- modelling
- processing
- quantitative analysis
- recognition
- recognition technology
- recognizer
- speech recognition
- speech recognition technology
- summarization
- transcription
Other assigned terms:
- annotated corpus
- annotation
- broadcast news
- case
- community
- content words
- continuous speech
- conversation
- conversational material
- corpora
- discourse
- document
- error rate
- general vocabulary
- generation
- knowledge
- language model
- language models
- language processing research
- language usage
- large corpus
- linear algebra
- measures
- method
- natural language
- pauses
- perplexity
- process
- punctuation
- qualitative analysis
- retrieval task
- sentence
- sentence boundaries
- sentence level
- signal
- speech corpora
- speech data
- speech signal
- spontaneous speech corpora
- standard deviation
- style
- technology
- terms
- test data
- test set
- text
- textbook
- toolkit
- topics
- training
- training data
- training material
- transcript
- transcriptions
- transcripts
- trigram
- trigram language model
- vocabulary
- vocabulary size
- wide-band speech
- word
- word error rate
- word error rates
- word usage
- words
