Call Statistics > Audio Quality
If your organization is experiencing problems with the audio quality of Unified Messaging (UM) calls and voice mail messages, use the Call statistics report to help you understand what's causing the problems.
The audio quality of a call can be affected by factors that aren't covered in the reports. For example, if a Mailbox server is experiencing a heavy memory load or CPU load, users may report poor call quality, even though the reports show excellent audio quality.
To get more details about the audio quality for a row in the report, select the row and click Audio Quality Details.
Date and Time: The UTC date and time that the call statistics were captured.
UM dial plan: The dial plan for the calls included in the statistics.
UM IP gateway: The gateway that took the calls included in the statistics.
NMOS: The Network Mean Opinion Score (NMOS) score for the call. The NMOS score indicates how good the audio quality was on the call as a number on a scale from 1 to 5, with 5 being excellent.
The maximum NMOS possible for a call is dependent on the audio codec being used, and NMOS may not be available for very short calls that are less than 10 seconds long.
NMOS degradation: The amount of audio degradation of the NMOS score from the top value possible for the audio codec being used. For example, if the NMOS degradation value for a call was 1.2 and the NMOS reported for the call was 3.3, the maximum NMOS score for that particular call would be 4.5 (1.2 + 3.3).
Jitter: The average variation in the arrival of data packets for the call.
Packet loss: The average percentage of data packet loss for the selected call. Packet loss is an indication of the reliability of the connection.
Round trip: The average round-trip score, in milliseconds, for audio on the selected call. The round-trip score measures latency on the connection.
Burst loss duration: The average duration of packet loss during bursts of losses for the selected call.
Number of samples: The number of calls that were sampled to calculate the averages.