| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a speaking-speed conversion device capable of stably outputting voice data after speaking-speed conversion processing when the speaking-speed conversion processing is performed on the voice data of a voice signal in real time. <P>SOLUTION: The present invention relates to a voice signal processing device for performing speaking-speed conversion processing. The voice signal processing device includes: input buffer means for accumulating voice data of an input voice signal; speaking-speed conversion means for extracting a basic period from a voice signal waveform for a retrieval period of voice data accumulated in the input buffer means and performing speaking-speed conversion processing by using the voice signal waveform of the basic period; output buffer means for accumulating the voice data after the speaking-speed conversion performed by the speaking speed conversion means; voice data output means for outputting output voice data frame every output interval; and conversion processing control means for causing the input buffer means to start the speaking speed conversion processing by the speaking speed conversion means after the voice data equal to or more than the minimum accumulation period longer than the retrieval period is accumulated. <P>COPYRIGHT: (C)2013,JPO&INPIT |
