Constructing Modulation Frequency Domain-Based Features for Robust Speech Recognition.
Jeih-Weih HungWei-Yi TsaiPublished in: IEEE Trans. Speech Audio Process. (2008)
Keyphrases
- frequency domain
- speech recognition
- feature extraction
- spatial domain
- speech recognition systems
- noisy environments
- hidden markov models
- fourier transform
- power spectral
- speech synthesis
- language model
- cross correlation
- power spectrum
- speech signal
- automatic speech recognition
- speaker identification
- pattern recognition
- denoising
- speech recognition technology
- speech recognizer
- audio signal
- feature set
- feature vectors
- subband
- power spectra
- neural network
- feature space
- single channel
- signal processing
- filter bank