Seeing wake words: Audio-visual Keyword Spotting.
Liliane MomeniTriantafyllos AfourasThemos StafylakisSamuel AlbanieAndrew ZissermanPublished in: BMVC (2020)
Keyphrases
- audio visual
- keyword spotting
- printed documents
- multi modal
- speech recognition
- visual information
- handwritten documents
- hidden markov models
- n gram
- speech processing
- multimedia
- multi stream
- document images
- character recognition
- visual data
- optical character recognition
- audio features
- high dimensional
- text documents
- language model
- image database
- video sequences
- keywords
- machine learning