Data-driven generation of text balloons based on linguistic and acoustic features of a comics-anime corpus.
Sho MatsumiyaSakriani SaktiGraham NeubigTomoki TodaSatoshi NakamuraPublished in: INTERSPEECH (2014)
Keyphrases
- data driven
- acoustic features
- text generation
- linguistic information
- natural language text
- linguistic patterns
- text data
- speech signal
- speaker verification
- text mining
- music genre classification
- information extraction
- visual features
- music information retrieval
- audio features
- natural language
- information retrieval
- cross correlation
- multi modal
- non stationary
- natural language processing
- similarity measure
- multimedia