finding audio pattern offset

I am trying to programmatically find the offset and length of an audio pattern. it’s shown clearly on the spectrogram. Can anyone give some suggestions? Thanks.
audio-spectrogram.jpg

Does it show clearly in the waveform?
If not, does it show clearly in the waveform after applying a high-pass filter at say 800 Hz?

(It’s much easier to work with amplitudes than frequencies).