Video of part of a P-SB7 spirit box session during a recent CRPI home investigation. Several very good responses come through. Antenna is enabled so there is some radio bleed, but there are a few responses coming through that do no appear to be radio. I was using the 150/ms sweep rate, which translates to scanning 7.5 frequencies per second. One particular response in this session is in the same male voice, and lasts at least 6-7 frequencies, meaning that in order for it to be radio bleed the intelligent sentence in the same voice would have had to be broadcast over at least two different radio stations simultaneously (knowing that an FM radio station can get a little bit of bleed at 1 frequency step on either side. For example, 100.0 FM could get a tiny bit of bleed on 99.9 and on 100.1). This is why I never go slower than the 150 ms (7.5 frequency per second) sweep rate, and often use the 100 ms and (on the P-SB11) even a 50 ms (20 frequencies/sec) sweep rate.