Release notes

Telnyx Voice AI agents now give you full control over response timing with the new Start Speaking Plan. Configure context-aware wait times for punctuation, pauses, or number sequences so your agent responds naturally and never interrupts users mid-sentence.

Telnyx RCS now supports the message.read webhook, giving businesses real-time read receipts and complete visibility into message engagement.

Telnyx has deployed GPUs in Sydney, enabling ultra-low-latency, onshore Voice AI Agents that deliver natural, human-like conversations for Australian businesses.

We’ve added built-in geolocation support for emergency calls (911/933). Devices can now pass latitude and longitude in SIP headers, and Telnyx automatically constructs a PIDF-LO and delivers it to emergency services.

Improving accuracy, standardization, and visibility across porting workflows

Automate everyday workflows in Jira, HubSpot, Salesforce, and more with real-time voice AI Agents.

We’ve expanded our Deepgram integration to support both TeXML and the Voice API, so you can add high-performance speech-to-text to any voice workflow. This builds on top of our existing Deepgram integration and Telnyx AI Agent integration.

Telnyx Voice AI Agents now support configurable background audio, letting you add ambient noise for more realistic experiences.

Telnyx Voice AI Agents now support warm transfers, allowing real-time handoffs between agents with conversation context included to avoid repeated information and improve call flow efficiency.

We’ve made some solid upgrades to media streaming that give you more flexibility when working with real-time audio, especially when you’re connecting to AI platforms or anything that expects audio in a specific format.This change removes a common integration headache and eliminates the need for additional audio transcoding. It simplifies your setup and cuts down on latency, complexity, and resource usage.

Telnyx now supports GPT-5 and Groq gpt-oss-120b for Voice AI Agents, delivering more choice, faster responses, and cost-efficient performance for real-time voice.