| 摘要 |
Cepstrum calculating means (5) obtains a cepstrum of a voice signal and mean-value calculation means (7) makes equal the cepstrum output. Threshold setting means (10) sets a voice detection threshold level on the basis of the cepstrum mean-value output. A cepstrum addition section (8) adds cepstrum value exceeding the cepstrum mean-value. A comparator (9) compares the cepstru output from the cepstrum addition section (8) with the threshold output signal from the threshold setting means (10), thereby to output voice-detected signal. |
