Screening for high risk suicidal states using mel-cepstral coefficients and energy in frequency bands.
Hande Kaymaz-KeskinpalaThaweesak YingthawornsukD. Mitch WilkesRichard G. ShiaviRonald M. SalomonPublished in: EUSIPCO (2007)
Keyphrases
- high risk
- cepstral coefficients
- frequency band
- frequency ranges
- energy distribution
- speech recognition
- speech signal
- linear prediction
- feature set
- subband
- audio signal
- high frequency
- low frequency
- wavelet transform
- hidden markov models
- risk factors
- prostate cancer
- feature vectors
- wavelet packet
- transfer function
- automatic speech recognition
- neural network
- multiresolution
- language model
- machine learning
- denoising
- risk assessment
- speaker recognition
- frequency domain
- quadtree
- filter bank