Fusing Various Audio Feature Sets for Detection of Parkinson's Disease from Sustained Voice and Speech Recordings.
Evaldas VaiciukynasAntanas VerikasAdas GelzinisMarija BacauskieneKestutis VaskeviciusVirgilijus UlozaEvaldas PadervinskisJolita CicelienePublished in: SPECOM (2016)
Keyphrases
- audio features
- feature set
- audio visual
- emotion recognition
- mel frequency cepstral coefficients
- acoustic features
- voice activity detection
- cepstral coefficients
- feature selection
- text to speech
- speech recognition
- multi modal
- speaker identification
- feature space
- spontaneous speech
- noisy environments
- low level
- feature vectors
- audio recordings
- visual information
- speaker verification
- prosodic features
- classification accuracy
- audio stream
- visual features
- audio signal
- feature extraction
- visual data
- feature types
- multimedia
- speaker recognition
- facial expressions
- video recordings
- speech synthesis
- feature selections
- music information retrieval
- broadcast news
- linguistic features
- image processing
- audio signals
- individual features
- structural features
- classification models
- feature subset
- training set