A video or voice Codec does not look or care about DSCP markings within the header as you suspected. The DSCP values within the Packet header inform routers down-stream that these packets should be given a pre-defined quality of service for buffering. Codec's strip away all the headers after of course noting the IP protocol type and payload type so the Codec knows how the payload within the packet is formatted and delivers this INFO to the codec during initial setup.
Often the packet payload is handed off to the codec processor (or DSP) at the beginning of the video call where the processor and codec are setup with all the specifics about the packet payload and how to decode the data. It is the payload type within the RTP header which determines the codec format. Payload Type - 112 is our CTS TelePresence dynamic payload type which tells the codec how the packet payload is formatted and how to decode the data.
Real-Time Transport Protocol
10.. .... = Version: RFC 1889 Version (2)
..0. .... = Padding: False
...0 .... = Extension: False
.... 0000 = Contributing source identifiers count: 0
0... .... = Marker: False
Payload type: DynamicRTP-Type-112 (112)
Sequence number: 28021
Timestamp: 9420000
Synchronization Source identifier: 0x00007777 (30583)
Payload: CFD505D7000000010A575D05000C03010A575D9BCC000401...
