Skip to main content
Jitter buffering smooths out packet arrival variation on SIP connections to reduce audio artifacts such as choppy or distorted speech. Enable this feature at the connection level to improve call quality on networks with inconsistent latency.

​
How it works

An adaptive jitter buffer temporarily holds incoming voice packets before playing them out, compensating for uneven packet arrival times (jitter). The buffer dynamically adjusts its size between configurable minimum and maximum values based on observed network conditions — expanding when jitter increases and shrinking when the network stabilizes.

​
Configuration settings

SettingPurposeDefaultRange
enable_jitter_bufferToggle jitter buffering on/offfalse—
jitterbuffer_msec_minMinimum buffer size (ms)6040–400
jitterbuffer_msec_maxMaximum buffer size (ms)20040–400
jitterbuffer_msec_min cannot exceed jitterbuffer_msec_max. The API will reject requests where the minimum is greater than the maximum.

​
Configuration via API

​
Credential connections

PATCH /v2/credential_connections/ 
{
  "jitter_buffer": {
    "enable_jitter_buffer": true,
    "jitterbuffer_msec_min": 60,
    "jitterbuffer_msec_max": 200
  }
}

​
FQDN connections

PATCH /v2/fqdn_connections/ 
{
  "jitter_buffer": {
    "enable_jitter_buffer": true,
    "jitterbuffer_msec_min": 60,
    "jitterbuffer_msec_max": 200
  }
}

​
IP connections

PATCH /v2/ip_connections/ 
{
  "jitter_buffer": {
    "enable_jitter_buffer": true,
    "jitterbuffer_msec_min": 60,
    "jitterbuffer_msec_max": 200
  }
}

​
Tuning guidance

  • Higher values increase latency tolerance and work better for high-jitter networks (e.g., international routes or mobile carriers).
  • Lower values reduce latency and are suited for stable, low-jitter networks (e.g., dedicated fiber or local connections).
  • Default values (60–200 ms) work well for most deployments and are a good starting point.
Jitter buffer configuration is currently available via API only. Portal UI support is coming soon.

​
Best practices

  1. Start with defaults before tuning — the default range of 60–200 ms handles most network conditions
  2. Monitor call quality metrics after enabling to validate the impact on your specific traffic
  3. Increase the maximum for routes with known high jitter rather than raising the minimum, which adds baseline latency to all calls
  4. Enable on specific connections rather than globally if only certain routes experience jitter issues
  5. Test during peak hours when network congestion and jitter are most likely to occur