Total average MOS Score for conversations

[Momoka_Tachibana]

Hi,

Is there any way that can extract total average "MOS "score for interactions?
I can see each score under the interaction, but I would like to know total average score as well.

[tim.smith]

You can use a conversations detail query to retrieve the MOS score dimensions and then average them.

[Momoka_Tachibana]

Thank you for the info,
I could extract MOS data using ```
"dimension": "mediaStatsMinConversationMos",

However, I am not sure which part of the metrics we should refer to because each interaction has 8 or 10 mos scores. 
Could you please give us information about what are these and which score we should refer to. Also, as long as checking MOS section through interaction view on PureCloud UI, the last of Metric(minMos) seems to be reflected.
Is there any reason for this? 


"mediaEndpointStats": [
                {
                  "codecs": [
                    "audio/opus"
                  ],
                  "minMos": 4.877959728240967,
                  "minRFactor": 92.25079345703125,
                  "maxLatencyMs": 39,
                  "receivedPackets": 1699
                },
                {
                  "codecs": [
                    "audio/opus"
                  ],
                  "minMos": 4.868839263916016,
                  "minRFactor": 91.8643798828125,
                  "maxLatencyMs": 30,
                  "receivedPackets": 193
                },
                {
                  "codecs": [
                    "audio/opus",
                    "audio/opus",
                    "audio/opus"
                  ],
                  "minMos": 4.878579616546631,
                  "minRFactor": 92.27748107910156,
                  "maxLatencyMs": 30,
                  "receivedPackets": 5579
                },
                {
                  "codecs": [
                    "audio/opus"
                  ],
                  "minMos": 4.878493785858154,
                  "minRFactor": 92.2737808227539,
                  "maxLatencyMs": 30,
                  "receivedPackets": 5413
                },
                {
                  "codecs": [
                    "audio/opus"
                  ],
                  "minMos": 4.878077507019043,
                  "minRFactor": 92.255859375,
                  "maxLatencyMs": 29,
                  "receivedPackets": 371
                },
                {
                  "codecs": [
                    "audio/PCMU"
                  ],
                  "minMos": 4.394517421722412,
                  "minRFactor": 92.46055603027344,
                  "maxLatencyMs": 30,
                  "receivedPackets": 25992
                },
                {
                  "codecs": [
                    "audio/opus",
                    "audio/opus",
                    "audio/opus"
                  ],
                  "minMos": 4.870382308959961,
                  "minRFactor": 91.92896270751953,
                  "maxLatencyMs": 31,
                  "receivedPackets": 18130
                },
                {
                  "codecs": [
                    "audio/opus"
                  ],
                  "minMos": 4.878614902496338,
                  "minRFactor": 92.27899169921875,
                  "maxLatencyMs": 32,
                  "receivedPackets": 13698
                }

[tim.smith]

Those stats for the participant should match up with the segments for the participant. Which item(s) you should use depends on which segment(s) you're interested in.

[Momoka_Tachibana]

Hi Tim-san,

Thank you for the info,
My understanding is the MOS score is measured between Endpoint(customer, agent) and Edge.
We would like to know how Agent's voice quality customer does receive. Measuring agent's audio quality from Customer side by using MOS is to know how customer take the voice quality of the agent.
In that case, which "purpose" (customer, agent)should I refer to?

[tim.smith]

Each participant has their own connection to PureCloud and therefore has their own mos scores. Look at the data on the participant you're interested in.

[Momoka_Tachibana]

If I am interested in how agent's voice qualiy does customer received,
I will look at the each MOS segments through Participant "agent"
Correct me if I am wrong.

Thank you

[Becky_Powell]

Questions about the base product should be posted to the PureCloud Community Forum or open a case with PureCloud Care. Thank you!