Blind speech segmentation using spectrogram image-based features and Mel cepstral coefficients.
Adriana StanCassia Valentini-BotinhaoBogdan OrzaMircea GiurgiuPublished in: SLT (2016)
Keyphrases
- cepstral coefficients
- speech signal
- speech recognition
- feature set
- linear predictive
- linear predictive coding
- automatic speech recognition
- linear prediction
- speaker identification
- mel frequency cepstral coefficients
- audio signal
- hidden markov models
- image segmentation
- non stationary
- spectral analysis
- speaker recognition
- noisy environments
- feature space
- pattern recognition
- multiscale
- feature vectors
- feature selection
- image features
- computer vision