Scale-invariant MFCCs for speech/speaker recognition.
Zekeriya TüfekciGökay DiskenPublished in: Turkish J. Electr. Eng. Comput. Sci. (2019)
Keyphrases
- speaker recognition
- scale invariant
- gaussian mixture model
- speaker verification
- speaker identification
- natural images
- affine invariant
- vector quantization
- keypoints
- scale space
- probabilistic neural network
- multiscale
- speech signal
- mel frequency cepstral coefficients
- speech recognition
- local binary pattern
- image structure
- invariant features
- texture images
- higher order
- image sequences
- pattern classification
- detection method
- mixture model
- image compression
- automatic speech recognition
- denoising
- feature extraction
- neural network