摘要 |
PROBLEM TO BE SOLVED: To improve the service and functions by extracting formants of a singing voice and estimating the sex and age thereby controlling the service and a KARAOKE performmance. SOLUTION: The singing voice 10 inputted from a singing voice input part 1 is inputted to a formant extraction part 2. The formant extraction part 2 cuts the vowels out of the inputted singing voice 10 to extract formants 11 by the vowels. The extracted formants 11 are inputted to a formant comparison part 3. The formant comparison part 3 compares the formants 11 with reference 12 to find the degree of the distinction of sex and estimated age 13 of the singing voice. The found degree of the distinction of sex and estimated age 13 are inputted to a KARAOKE performance control part 7 and inputted to a fortune data base 5. Then the KARAOKE performance control part 7 controls the KARAOKE performance part 8 in real time according to the degree of the distinction of sex and the estimated age 13 inputted momentarily from the formant comparison part 13. |