I don't think I'm seeing that on my 12.5 system (using IPC and a 7975) (see attached).
Via modifying the makeCall sample (note, observing only the near-side address and terminal:)
12:12:46.27 Holding call: (P1-dstaudt) GCID=(1,3009)->ACTIVE
Received--> Terminal/(P1-dstaudt) [SEP001EF727852D] CiscoRTPInputStoppedEv [#23] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12
Received--> Terminal/(P1-dstaudt) [SEP001EF727852D] CiscoRTPOutputStoppedEv [#24] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12
12:12:46.59 Sleeping 5 seconds...
Received--> Call/(P1-dstaudt) 3009/1 CallCtlTermConnHeldEv SEP001EF727852D [#25] Cause:100 CallCtlCause:100 CiscoCause:0 FeatReason:12
12:12:51.59 Resuming call: (P1-dstaudt) GCID=(1,3009)->ACTIVE
12:12:51.83 Sleeping 5 seconds...
Received--> Call/(P1-dstaudt) 3009/1 CallCtlTermConnTalkingEv SEP001EF727852D [#26] Cause:100 CallCtlCause:214 CiscoCause:0 FeatReason:12
Received--> Terminal/(P1-dstaudt) [SEP001EF727852D] CiscoRTPOutputStartedEv [#27] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12
Received--> Terminal/(P1-dstaudt) [SEP001EF727852D] CiscoRTPInputStartedEv [#28] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12
Are the events from different terminal connections? You should get a set for each one in the call if you're observing both ends...
Possibly two (identical/redundant) observers are in place?
If you need a detailed analysis, I would suggest opening a ticket with DevNet Developer Support, who can take a look at the detailed logs to see what is happening: https://developer.cisco.com/site/support/