@Edward Gonzales That's likely because Genesys and Cyara are using different MOS methodologies. Even if they're measuring the same call, E-MOS and the MOS reported by Genesys may weight factors such as codec, packet loss, jitter, and latency differently.
It would be interesting to compare the raw metrics for the same call (packet loss, jitter, latency, codec, and call leg) rather than just the final MOS value. That should help identify where the scores are diverging.
I also came across a Cyara resource that may provide additional insight into how their voice quality measurements are calculated:
(Datasheet: Voice Quality Measurement & Transcoding Detection with Cyara Voice Assure - Cyara)
------------------------------
Akhil
------------------------------
Original Message:
Sent: 09-04-2026 16:09
From: Edward Gonzales
Subject: MOS Score algorithm
thanks all. Now to find out why Cyara gives me a different score for the same call.
------------------------------
Edward
------------------------------
Original Message:
Sent: 09-04-2026 05:53
From: Akhil Akhil
Subject: MOS Score algorithm
Hi @Edward Gonzales,
According to the Genesys documentation, MOS uses an industry-standard methodology to rank audio quality from 1 (unacceptable) to 5 (excellent).
For calls with multiple legs (station, edge-to-edge, and PSTN), Genesys Cloud uses the lowest MOS as the interaction score.
However, the documentation does not provide the detailed calculation logic or formula used to derive the MOS.
Mean Opinion Score (MOS) overview - Genesys Cloud Resource Center
------------------------------
Akhil
------------------------------