Overlap-add Windows with Maximum Energy Concentration for Speech and Audio Processing.
Tom BäckströmPublished in: ICASSP (2019)
Keyphrases
- audio visual
- audio stream
- cepstral features
- high energy
- broadcast news
- multimedia
- minimum energy
- real time
- audio recordings
- audio signals
- emotion recognition
- speech recognition
- linear predictive coding
- digital audio
- audio video
- speech processing
- text to speech
- speaker identification
- multi modal
- energy consumption
- signal processing
- audio features
- information processing
- visual information
- energy minimization
- content based video retrieval
- human language
- multi stream
- speech signal
- human computer interaction
- maximum number
- acoustic signals
- low level
- spoken documents
- data processing
- dialogue system
- speech music discrimination