Missed the LIVEWire Webinar? We've Got You Covered.

If you couldn’t catch our live session on Why Voice AI Fails Before the First Call, the recording is now available!

In this session, we covered:

• Prompt-Stuffed Architecture: Why putting business logic inside prompts leads to inconsistent results and hallucinated behavior.

• Implicit State Machines: What happens when conversation flow lives inside an LLM instead of an explicit state model.

• Ignoring the Voice Latency Budget: Why a few seconds of silence destroys the caller experience and how to measure the full voice pipeline.

• Tool Calls as an Attack Surface: How giving LLMs too many responsibilities creates unpredictable system behavior.

• Zero Post-Call Observability: Why transcripts alone aren't enough to debug AI agents and what production telemetry actually needs to capture.

You’ll also see the architectural patterns used to solve these problems, including state-governed agents, constrained tool calls, and real observability for AI-driven voice systems.

▶️ Watch the recording now and start building smarter!

Explore this week’s updates and releases! 🔎

Error Details Now Visible in Message Logs

Message logs now display error codes and error messages directly in the UI, giving you full visibility into failed messages without needing to make a separate API call.

Previously, this information was only accessible via the API. Now, if either value is present, you can see it at a glance in the dashboard — making it faster to diagnose delivery issues without needing to make a separate API call.

Scripts Filter Split into SWML and CXML

The Resources UI filter for Scripts has been separated into two distinct filters: SWML and CXML.

This makes it easier to find and manage your scripts by type without sorting through a combined list.

Flexible Codec Format Support for Dial Requests

You can now pass codecs as either an array of strings or a comma-separated string when making a calling/calls dial request, giving you greater flexibility in how you configure your calls.

Supported codecs include: PCMU, PCMA, OPUS, G729, G722, VP8, and H264.

📖 See the full codecs parameter documentation for details.

Explore this weeks updates and releases! 🔎

Hold Music URL on SIP Endpoints

Physical SIP phones can now have a set hold music URL that will be played when hold is pressed on an outbound or inbound call to that SIP endpoint.

Explore this weeks updates and releases! 🔎

Conference Audience Feature Deprecation

The Conference Audience feature has been deprecated and removed as part of ongoing platform modernization efforts.

This change supports improvements to platform stability and enables new regional infrastructure.

Removal of Dashboard Info Panel

The Home Dashboard has a new look! To improve navigation, usability and access to resources the expandable info panel has been removed and you will now see a cleaner, text-based dashboard.

• New dedicated links for Changelog, Support, Docs, and Community resources can be found in the top right menu, Including direct access to the Status page and Documentation links ( that now open in a new window for easier reference)

⚠️ Shutdown of Bahrain RTC POP

The Bahrain RTC POP will be retired on March 27, 2026.

• Calls will automatically route to the next closest available node. No action needs to be taken at this time.

• Any active calls on this node will continue until completion.

Introducing a new Voice Analytics Dashboard, giving you a high-level view of voice call health across your project.

This new dashboard makes it easier to identify potential issues, monitor performance trends, and understand overall voice health — all without needing to build custom observability workflows.

Where to find the new dashboard.

The Voice Analytics Dashboard is now available. Navigate to “Analytics” from the left hand menu →  select the submenu “Voice”.

We want your feedback!

Once you have had time to use the new dashboard, we would love to hear your feedback! Take the Voice Analytics Dashboard Feedback Survey.

Explore this weeks updates and releases! 🔎

✨ Code Editor Toolbar Improvements

The code editor toolbar has been enhanced for easier script management.
You can now copy scripts directly from the toolbar, and read-only editors still provide access to tools like copy and resize.

🧪 New “Click to Test” Button

You can now test resources directly from the browser.
A new Click to Test button on the Resource detail screen lets you initiate a test call without assigning a handler or using an external phone.

Explore this weeks updates and releases! 🔎

🛠 Bug Fixes

Inbound Messages Now Visible in Logs
Inbound messages sent to numbers without a handler assigned will now appear in logs.
Why it matters: Improved troubleshooting and full message visibility.

Phone Number Status Callback Update Restored
Fixed an issue preventing status callback URLs from being updated when a Relay application handler was assigned.
Why it matters: Phone number configurations now update as expected.

FriendlyName Persists via Compatibility API
The optional FriendlyName field now stores correctly when creating numbers through the Compatibility API.
Why it matters: More reliable API configuration.

Explore this weeks updates and releases! 🔎

New Call Quality Metrics

We’ve added Call Quality Metrics to your call details view to help you quickly understand call performance and troubleshoot audio issues. You can now review:

• MOS score (overall audio quality)

• Jitter (audio stability)

• Packet loss & packet stats (network impact on audio)

• DTMF and advanced metrics (helpful for diagnosing keypad/IVR interactions)

These metrics are available on a per-call basis so you can pinpoint what happened and share details with your team faster.

Improved SWML Script Editor
We’ve unified the code editor experience across legacy Spaces and the Call Fabric UI. There is now one editor with improved usability and validation features, including:

• Formatting improvements (multi-line rendering and cleaner defaults)
• Syntax checks for YAML/XML (basic error detection)
• Schema validation for SWML (YAML)
• Resizable editor options
• Default starter content when creating a new SWML Script (no more blank editor)

Explore this weeks updates and releases! 🔎

Improved Campaign Setup Guidance

We’ve improved the tooltips in the managed campaign creation flow to make setup clearer and more intuitive. Tooltips that previously referenced default values now display example values, helping you better understand what inputs are expected as you configure your campaign.

Clearer Keyword Field Examples

We’ve also updated the keyword-related tooltips in the managed campaign creation form. These tooltips now provide example keywords instead of referencing defaults, reducing confusion and making campaign setup faster and easier.