Using Approximated Auditory Roughness as a Pre-Filtering Feature for Human Screaming and Affective Speech AED.
Di HeZuofu ChengMark Hasegawa-JohnsonDeming ChenPublished in: INTERSPEECH (2017)
Keyphrases
- pre filtering
- event detection
- emotional state
- language acquisition
- speech recognition
- emotion recognition
- signal processing
- speech signal
- text to speech
- feature vectors
- linear combination
- image features
- human behavior
- human subjects
- computational models
- automatic speech recognition
- cognitive model
- speech synthesis
- spoken dialogue systems
- human emotion
- working memory
- affective states
- audio visual
- fractal dimension
- human computer interaction
- hidden markov models