More Than 100% inlvl in DSD from GQRX UDP

[Tanzanite]

Hi,

I output the audio from GQRX to DSD for decoding using UDP. nc -l -u localhost 7355 | dsd -i- -o pa:2 command is used to launch DSD and pipe in the UDP data. The connection is working fine and I am listening to a DMR amateur radio station. However, I found out that the inlvl is usually more than 100% which is not desirable from the DSD wiki operation (Operation). The wiki suggests typical values of 40% for C4FM and 66% for QPSK and indicates to change the input gain using Datascope mode. However, I am not sure how to achieve this. An excerpt of the program output is shown below.
Also, the voice I heard is a short period of sound for like half a second, and the most circumstances are the Unknown burst type: 1101. I think that the more than 100% inlvl issue is affecting the voice decoding.

Sync: +DMR o: 109 mod: GFSK g: 25.000000 inlvl: 104% slot0 [slot1] Unknown burst type: 1101
Sync: +DMR o: 890 mod: QPSK g: 25.000000 inlvl: 102% slot0 [SLOT1] VOICE e:
Syncing voice output stream
Start voice output stream
========
Sync: (+DMR) o: 24 mod: QPSK g: 39.449997 inlvl: 108% [slot0] SLOT1 VOICE e:=====R=======R===========R==========R==========R=====R==
Sync: +DMR o: 166 mod: C4FM g: 11.732796 inlvl: 107% slot0 [slot1] Unknown burst type: 1101
Sync: no sync
Sync: +DMR o: 108 mod: GFSK g: 25.000000 inlvl: 105% slot0 [slot1] Unknown burst type: 1101
Sync: +DMR o: 1183 mod: QPSK g: 25.000000 inlvl: 113% slot0 [SLOT1] VOICE e:
Syncing voice output stream
Start voice output stream

Sync: (+DMR) o: 24 mod: QPSK g: 50.000000 inlvl: 116% [slot0] SLOT1 VOICE e:=====R===========================E====R====R=======R====R===
Sync: +DMR o: 166 mod: C4FM g: 27.576511 inlvl: 112% slot0 [slot1] Unknown burst type: 1101
Sync: no sync
Sync: +DMR o: 109 mod: GFSK g: 25.000000 inlvl: 101% slot0 [slot1] Unknown burst type: 1101