Syllable news and updates

Syllable release 26.10.20 adds a Tester tab that checks a saved Insights tool configuration against past sessions or uploaded files before it goes live, and prompt caching so several Insights tools can reuse shared conversation context. Web Search tools now fall back to OpenAI when the primary search provider fails, so agents keep answering from the websites they are configured to use.

Our read

Testing on past calls catches configuration mistakes before patients hear them, and caching cuts repeated model costs. The fallback adds a provider to the chain, which Syllable's trust center lists among its subprocessors through Azure OpenAI.

Syllable release 26.10.15 retired Sheet Insights workflows and the Sheet AI and eValidator tools, which no longer run, though existing configurations and history stay for reference. The release also improved the warnings that flag configurations using models due to be retired across prompts, campaigns and insights workflows.

Our read

Teams that relied on the Sheet tools need to move to Syllable's supported workflows. The model warnings help keep retired models out of new campaigns.

Syllable release 26.10.1 adds five Gemini 3.x models to its prompt options and turns off the temperature setting for GPT-6 Sol and Luna to match gateway support. It also adds deprecation warnings for older language models and for Google speech to text V1, and it retries empty model tool responses immediately to make Insights more reliable.

Our read

Deprecation warnings give agent builders notice before a retired model breaks a live workflow. That matters more as model options in these platforms now change week to week.

Syllable release 26.9.29 adds GPT-6 Sol and GPT-6 Luna to the OpenAI and Azure OpenAI models available to its voice agents. The release also fixes an edge case that caused duplicate outbound campaign calls, improves voice routing during temporary service errors, and adds billing and usage data to Airbyte data feeds.

Our read

The duplicate call fix matters most, since repeated outbound calls reach patients directly. Usage data in Airbyte lets teams track model costs alongside the rest of their reporting.

Syllable release 26.9.22 added Insights Reports, a workflow detail view for reviewing analyzed results, configuration and session context, and Experiments, which splits live traffic across weighted voice agent variants and compares the results by session. The release also fixed voicemail delivery for carriers with short recording windows.

Our read

Experiments turn agent changes into measured tests on real calls. Health systems should agree on the outcome being tested, such as completed scheduling, and on how patients are informed, before splitting live patient traffic between variants.