
Enhanced speech recognition model is now available
62% Word Error Rate (WER) improvement for US English

62% Word Error Rate (WER) improvement for US English

Following Google’s release of new Speech API, we are happy to announce improved quality of call records transcription.

Voximplant now includes a native Deepgram module that connects any Voximplant call to Deepgram’s Voice Agent API for real-time, speech‑to‑speech conversations. You can stream audio from phone numbers, SIP trunks, WhatsApp, or WebRTC into Deepgram’s unified agent environment—combining STT, LLM reasoning, and TTS—and play responses via Voximplant’s serverless runtime with minimal latency.

New Features in Voximplant Kit: Update overview We are constantly working to improve our product to make it easier to use and more effective for you. In this update, we have added several useful features. Here’s what’s new:

Learn how a Voice AI Orchestration Platform connects LLMs, STT/TTS, turn‑taking, and telephony (PSTN, SIP, WebRTC) to build reliable real‑time voice agents. See benefits, architecture, and how Voximplant helps.

OpenAI has recently announced GA version of their Realtime API that Voximplant now fully supports

New integrations for Voice AI have arrived: Google's Gemini 2.0 Flash model, featuring seamless voice-to-voice conversation capabilities and ElevenLabs low-latency streaming speech synthesis are now available for Voximplant developers

The new integration enables instant connection of any Voximplant call to an Ultravox agent, delivering seamless voice-to-voice conversations.

Connect any Voximplant call to ElevenLabs Conversational AI agents

Check out the latest useful Voximplant Kit updates — we developed chat analytics, improved call history, added new tools for supervisors, expanded scenario capabilities, and updated the softphone. Below is a brief overview of the essential enhancements.