Voximplant

Two months ago we launched AI Agent Skills. Since then we've received a lot of positive feedback from you, telling us that building voice AI agents on Voximplant has gotten a lot easier and faster To make that process even smoother, we recorded a full walkthrough. Link in the comments ⬇️ Watch Claude take an ElevenLabs voice agent, deploy it on Voximplant, and place a live call, all without a single line of code written by hand. The whole process takes about 15 minutes We're planning similar walkthroughs for other popular AI agents. Let us know which one you want to see next

Voice AI pilots stall for one reason more than any other: the client needs compliance answers covering every provider in the stack before they can proceed Every component in a voice AI pipeline is a potential data transfer. Each one needs an answer, and if you can't give it for even one provider, the conversation stops there Voximplant gives you the tools and infrastructure to answer those questions across the stack. The telephony layer supports EU data residency and comes with compliance documentation. For LLM layer we added baseUrl support to give you control over where the requests go Here’s a full guide to running a sovereign voice AI stack on Voximplant: https://lnkd.in/d-Z5Xgfs Not sure if Voximplant fits your compliance requirements? Reach out, we are happy to help

Kindly reminder that we have a Discord community. It's relatively small right now, but genuinely cozy A few reasons to hang out there: — get unstuck faster than waiting on a ticket — catch feature updates early — compare notes with other developers solving similar problems — have a say in what we build next We'd rather keep growing this with people who actually stick around rather than chase numbers If that sounds like your kind of place, come find us: https://lnkd.in/dGDgtrK4

In a speech-to-speech pipeline call audio goes directly to the model. That makes the question of where data is processed more immediate than with text-based LLMs (we covered them in the previous post: https://lnkd.in/dzYU-ffq⁠) VoxEngine's Realtime API client accepts a baseUrl parameter, the same way Chat Completions and Responses API do. Point it at any OpenAI Realtime-compatible endpoint and that's where audio goes For EU data residency Azure OpenAI is a working example. Deploy a Realtime model in an EU region, set baseUrl to your Azure resource endpoint, and call audio stays in that region. Other providers with regional Realtime deployments work the same way. Just make sure to check the deployment type! Full article: https://lnkd.in/daM6PUnm Docs with Azure scenario: https://lnkd.in/deQ3QEgV This is the second post in a series on building compliant voice AI on Voximplant. Next is a complete guide to running a fully sovereign voice AI stack end to end Found this useful? Let us know with a reaction

A voice agent that sounds natural but drifts from its script isn't production-ready. Disclaimers, prices, compliance lines need to be said exactly OpenAI's new gpt-realtime-2.1 addresses that directly: better alphanumeric recognition, better silence and noise handling, and improved interruption behavior, at the same price as 2.0 Alongside it, gpt-realtime-2.1-mini: a distilled model built for realtime voice, faster than the flagship 2.1 and roughly three times cheaper Both are already supported in VoxEngine. We tested it against inbound, outbound, function calling, and hybrid pipeline scenarios, and it holds up great. Definitely worth trying More in the docs: https://lnkd.in/dFKnWggm

Not every voice AI platform makes it easy to control where your LLM requests go. We do In VoxEngine Chat Completions and Responses API both accept a baseUrl parameter. Point it at any OpenAI-compatible provider or your own deployment, and that's where requests go. You choose which model handles your calls, and where it runs Which API option to use depends on your chosen LLM and what your provider supports Keep in mind that a call transcript may contain personal data, and it goes wherever the LLM endpoint is hosted. If your clients have data localization requirements, baseUrl lets you control where requests are processed You can find out more in our blog post: https://lnkd.in/dncSXEGB Or check the documentation with examples: https://lnkd.in/dEKJAzgf This is the first post in a series on building compliant voice AI on Voximplant. More on running a fully sovereign voice AI stack is coming What would you like to know? Drop your questions in the comments

Realtime voice agents have become more intelligent and responsive, but many still struggle with how the agent actually sounds. They can answer correctly and still feel flat or disconnected from the caller's tone Inworld's Realtime API is built to address that. It gives agents more expressive delivery, conversational awareness, and consistent voice identity across 200+ languages Voximplant now supports it natively, so you can bring those agents into real phone calls, SIP networks, and WhatsApp without custom media infrastructure. After all, Voice AI is only as good as the options you have to build with Give it a try, we’d love to hear your thoughts! Demo: https://lnkd.in/ezHx5eY2 Blog post: https://lnkd.in/ecPKaagv Docs: https://lnkd.in/ejXWHp68

We’ve published Voximplant AI Agent Skills Now agents like Cursor, Claude Code, or Codex can write VoxEngine scenarios, automate platform setup, deploy and test call flows, and debug real calls from logs This is part of a broader direction for us: making Voximplant agents-friendly. We believe that as AI coding tools become a standard part of how developers work, the platforms they build on should work just as naturally with those tools All you need is a Voximplant account and a service account for access. Your AI agent will walk you through the setup Try it out and share your feedback: https://lnkd.in/dFJQnNiD

Voice AI infrastructure should give developers control, not take it away Today we’re shipping a native MCP Client in VoxEngine. Connect to any MCP server directly from your scenario code. No gateway, no middleware, no extra service to maintain. The same tool layer works across providers: switch models without rebuilding your integrations You can now have full visibility over every tool call, the ability to add telephony context, adapt parameters, handle fallbacks — all in the same place you manage the rest of the call Demo: watch an agent create a live Zapier request before the caller hangs up → https://lnkd.in/dJ3KvDQf Docs: check the setup guide, architecture, and full scenario code → https://lnkd.in/d5SwK4SB Blog post: read more → https://lnkd.in/d82ZwidC

New: Voximplant Secrets — secure credential storage for VoxEngine Modern Voximplant scenarios authenticate against a long list of third-party services: Voice AI providers, STT and TTS vendors, CRMs, payment systems, and more. Today we're giving each of those credentials a dedicated home with Secrets — native credential storage for VoxEngine. What's included: - Up to 8,192 bytes per secret — large enough for credentials like GCP's credentials.json - Single-call retrieval inside any scenario: VoxEngine.getSecretValue("NAME") - Manage interactively from the Voximplant portal, or programmatically via the Management API - Role-gated access — only Administrator, Owner, and Developer roles can manage secret values - CI/CD-friendly through the Management API Generally available today across all Voximplant regions, at no additional cost. Read the full announcement → https://lnkd.in/dB7cByx2 #DeveloperTools #CPaaS #CICD #VoxEngine #API