WebRTC Softphone Audio Dropouts on AP-Southeast-1 BYOC Trunks

Quick question about the correlation between WebRTC softphone audio stability and our specific carrier failover configurations in the AP-Southeast-1 region. We are currently managing 15 Bring Your Own Carrier (BYOC) trunks, and while SIP registration and outbound routing remain stable, agents utilizing the Genesys Cloud WebRTC softphone are reporting intermittent audio dropouts specifically during peak hours. This issue appears isolated to the AP-Southeast-1 deployment and does not affect our US-East or EU-West regions.

The environment consists of Genesys Cloud version 2024.10 with the latest softphone client updates deployed across all agent workstations. Network diagnostics indicate that jitter remains within acceptable thresholds (<30ms) and packet loss is negligible (<0.1%) when tested directly against the carrier gateways. However, once the call is bridged through the Genesys Cloud media servers, the WebRTC stream exhibits significant degradation. We have verified that the SIP credentials and outbound routing rules are correctly configured, and the trunks report a ‘Healthy’ status in the administration console.

Upon reviewing the detailed call logs and media server metrics, we noticed a slight increase in CPU utilization on the media servers during these dropout events. The error logs do not show explicit SIP errors, but there are occasional warnings related to codec negotiation failures between the softphone client and the carrier gateway. We are using the G.711 codec exclusively for these trunks to minimize transcoding overhead, yet the issue persists. The failover logic seems to trigger unnecessarily, causing brief disconnections before re-establishing the connection on the same primary trunk.

Has anyone encountered similar issues with WebRTC softphone audio quality when using BYOC trunks in the AP-Southeast-1 region? We are considering adjusting the jitter buffer settings or implementing a different codec strategy, but we want to ensure that the root cause is not related to carrier-specific quirks or media server configuration. Any insights into optimizing WebRTC performance for high-volume BYOC deployments would be greatly appreciated.