| 摘要 |
A signal processing device (9) samples an audio waveform (eg. speech) and generates a new waveform by extracting samples (eg. quiet or non-speech portions of the waveform, fig. 4A) and combining the remaining samples in a stretched waveform (fig. 4B, eg. via linear interpolation) before outputting the new waveform through a speaker (7). The stretched waveform may achieve the same overall duration as the original but with enhanced intelligibility for eg. aircraft communications. |
