Linear Predictive perceptual filtering for Acoustic Vector Sensors: Exploiting directional recordings for high quality speech enhancement.
Muawiyath ShujauChristian H. RitzIan S. BurnettPublished in: ICASSP (2011)
Keyphrases
- acoustic features
- speech enhancement
- speech signal
- linear predictive
- high quality
- mel frequency cepstral coefficients
- noisy environments
- single channel
- speaker verification
- speech recognition
- automatic speech recognition
- vocal tract
- linear prediction
- speaker identification
- multi channel
- noise reduction
- hidden markov models
- sound source
- signal to noise ratio
- spectral analysis
- non stationary
- music information retrieval
- speaker recognition
- high resolution
- visual features
- human visual system
- image quality
- frequency domain