A spectral autocorrelation method for measurement of the fundamental frequency of noise-corrupted speech.
M. LahatRussell J. NiederjohnDavid A. KrubsackPublished in: IEEE Trans. Acoust. Speech Signal Process. (1987)
Keyphrases
- fundamental frequency
- cost function
- computational cost
- computational complexity
- high precision
- data sets
- higher order
- hybrid method
- segmentation method
- mutual information
- experimental evaluation
- significant improvement
- pairwise
- similarity measure
- detection method
- error rate
- multispectral images
- spectral analysis
- machine learning