Double-talk detector based on speech feature extraction for acoustic echo cancellation.
Mahfoud HamidiaAbderrahmane AmrouchePublished in: SoftCOM (2014)
Keyphrases
- feature extraction
- speech sounds
- acoustic features
- speaker identification
- mel frequency cepstral coefficients
- speech recognition systems
- speech recognition
- speaker independent
- speech signal
- automatic speech recognition
- emotional speech
- feature vectors
- prosodic features
- pattern classification
- acoustic signal
- image processing
- speech synthesis
- acoustic models
- linear predictive
- image classification
- speaker recognition
- speech segments
- speech recognizers
- preprocessing
- cepstral features
- audio visual
- audio signal
- face recognition
- cepstral coefficients
- source localization
- feature selection
- linear feature extraction
- formant frequencies
- discriminant analysis
- dimensionality reduction
- spoken language
- broadcast news
- feature extractors
- speech recognizer
- vowel phonemes
- pattern recognition
- feature set
- principal component analysis
- extracted features
- emotion recognition
- feature space
- vocal tract
- gabor filters
- texture features
- underwater vehicles
- hidden markov models
- visual features
- text to speech
- noisy speech
- speaker verification
- detection algorithm
- detection method
- audio stream
- face images
- wavelet transform
- multiresolution
- support vector machine svm
- non stationary
- spectrum analysis
- bird species
- high frequency
- gaussian mixture model
- texture analysis
- noisy environments
- audio features