| 摘要 |
<p>PROBLEM TO BE SOLVED: To provide a voice processing device, a method and a program capable of properly complementing a voice component lost in an arbitrary frequency band.SOLUTION: A voice processing device includes: an extraction unit; a detection unit; a generation unit; a conversion unit; and a complement unit. The extraction unit extracts a voice parameter representing a voice component for each fragmented frequency band from a spectral envelope of input voice. The detection unit detects a lost band, which is a frequency band where the voice component is lost, in the spectral envelope of the input voice. The generation unit generates the voice parameter corresponding to the lost band, based on the position of the lost band, statistical information and the voice parameter extracted from the spectral envelope of the input voice. The conversion part converts the voice parameter corresponding to the lost band to the spectral envelope of the lost band. The complement unit synthesizes the spectral envelope of the lost band and the spectral envelope of the input voice and generates the spectral envelope where the lost band is complemented.</p> |
