Acoustically-Driven Phoneme Removal That Preserves Vocal Affect Cues.
Camille NoufiJonathan BergerMichael C. FrankKaren ParkerDaniel L. BowlingPublished in: CoRR (2022)
Keyphrases
- affective states
- data driven
- automatic speech recognition
- hidden markov models
- speech recognition
- computer vision
- emotion recognition
- context dependent
- speech signal
- speech synthesis
- acoustic features
- web services
- multi modal
- appearance cues
- visual cues
- real time
- object recognition
- three dimensional
- image segmentation
- image processing
- knowledge base
- learning algorithm