| 摘要 |
An audio signal time offset estimation algorithm estimates a time offset between two audio signals. The audio signal time offset estimation algorithm provides a way to measure that delay, even when the audio equipment causes severe distortion, and the signal coming out of the equipment sounds very different from the signal going in. Normalizing block algorithms provide perceptually consistent comparison of speech signals. These algorithms compare the sounds of two speech signals in a way that agrees with human auditory perception. This means, for example, that when these algorithms indicate that two speech signals sound identical, it is very likely that persons listening to those speech signals would describe them as identical. When these algorithms indicate that two speech signals sound similar, it is very likely that persons listening to those speech signals would describe them as similar.
|
