CEO & Co-founder of Voximplant, the world's first serverless cloud communications platform
Dubai, UAE
Joined July 2009
- Tweets4.1K
- Following822
- Followers894
- Likes555
We take speech-to-speech seriously: @alibaba_cloud has just released its latest s2s model, and we wasted no time making it available for our devs - giving them more options, especially in APAC region. #voiceai
⚡️Qwen-Audio-3.1-Realtime is now on Voximplant via a new QwenCloud provider. Full-duplex speech model with tool calling on your own phone numbers or SIP. Just bring your own QwenCloud API key.
go.voximplant.com/Pbcx5C
#VoiceAI #Voximplant #Qwen
Just use 'grok-voice-latest' in your call scenario and you are ready go go! #voiceai
🎙️xAI’s Grok Voice 3.0 has run on Voximplant since launch day, responding to callers with sub-second latency
A full inbound call scenario to start from: go.voximplant.com/mdVQmU
#VoiceAI #Voximplant #Grok
Agentic VoiceDSP has arrived: use turn detection, VAD, noise suppression in your VoiceAI calls using one API. More to come.
True speech-to-speech models and APIs are here to stay. Use Voximplant's GPT Live API Integration to launch your Voice AI agents in full-duplex mode and carry less of the burden when it comes to things like turn detection and response latency.
GPT-Live-1 by OpenAI is now live on Voximplant Platform
It keeps talking while it processes hard questions in the background, then folds the answer back in. And it is better at getting order numbers and confirmation codes right
Try it with us: go.voximplant.com/h5Ak38
Alexey Aylarov retweeted
Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking launched today and both are already running on Voximplant
No more lag mid-call or awkward silence while the agent looks something up
Huge congrats to @googledevs on the launch and thanks for trusting us with early access
Alexey Aylarov retweeted
Say "Hello, ¡Hola!, 你好" to Gemini 3.8 Live & 3.8 Live Extended Thinking. 🎙️⚡
Built for next-gen voice agents that reason, execute, and keep the conversation flowing in real time:
⚙️ Async function calling: Trigger tools in the background without pausing audio streams.
👁️ Live visual grounding: Contextualize responses based on what users see.
🎯 Alphanumeric precision: Reliably parse confirmation codes, claim IDs, and technical data.
🌍 Global reach: 97+ languages with natural accent consistency.
🔄 Incremental content updates: Seamlessly backchannel text-based context.
Ready to build agents that truly listen, reason, and act?
Great new release from Google for Voice AI, @voximplant 's Gemini Live API Integration does support it right now! Give it a try to talk with Google's most advanced models today
Say “hi” to our most advanced audio models from @GoogleDeepMind yet built for natural, production-ready voice applications.
🔷 Gemini 3.8 Live
🔷 Gemini 3.8 Live Extended Thinking
With these models, you can speak naturally, collaborate easily, and tackle complex tasks using just your voice.
.@Voximplant: 390M+ calls connected successfully in August alone. Scaling our Voice AI backbone to power conversations worldwide. 🚀
Alexey Aylarov retweeted
Two months ago we shipped Agent Skills for Voximplant. Since then we recorded a full walkthrough to make vibe coding AI agents even simpler
Watch Claude take an ElevenLabs voice agent, deploy it on Voximplant, and place a live call with zero manual code, ~15 min end to end 👇
Alexey Aylarov retweeted
Realtime API in VoxEngine accepts baseUrl. Point it at any OpenAI Realtime-compatible endpoint and that's where call audio goes
For EU data residency Azure OpenAI is a working example. Just make sure to check the deployment type before going to production
Data residency is a critical factor for Voice AI integrations, but with @Voximplant, you retain complete oversight and control over exactly where your data is stored and processed.
Voice AI pilots stall when clients ask where their data goes and you can't give answers
We provide you with tools and infrastructure to give those answers. Here's a full guide to running a sovereign voice AI stack: go.voximplant.com/3L6rTQ
The diagram with the full data flow ⬇️
Great to see @voximplant on the 1st place in the top 10 best WebRTC services of 2026 by @gitnux gitnux.org/service/webrtc/
Alexey Aylarov retweeted
OpenAI released gpt-realtime-2.1: better alphanumeric recognition, noise handling, and interruption behavior, same price as 2.0
Mini version is faster than the flagship and roughly three times cheaper
Alexey Aylarov retweeted
VoxEngine Chat Completions and Responses API work with any OpenAI-compatible text LLM
Set baseUrl to any provider or your own deployment
Useful when clients have data localization requirements and you need control over where call data is processed
Our integration with @inworld Realtime API is now live! Give it try!
Inworld Realtime API is now supported by Voximplant
Expressive, conversation-aware speech-to-speech on real phone calls, SIP, and WhatsApp without custom media infrastructure
Consistent voice identity across 200+ languages
Demo below ⬇️
youtu.be/1jQ9UZfCABQ?utm_sou…
Integration with MCP servers can be done using new MCP Client module, give it a try today #voiceai
Native MCP Client is now in VoxEngine — connect to any MCP server directly. The same tool layer works across OpenAI, Gemini, Grok, ElevenLabs and other LLM providers
Watch demo: agent creates a Zapier request during a live call
youtu.be/fyjSWILjyMQ
Vibe-coding your Voice AI agents just got a whole lot easier! #voiceai #vibecoding
Voximplant AI Agent Skills are out
Write scenarios and manage the platform directly from your agent. No more switching between your IDE, docs, and console
Works with Cursor, Claude Code, Codex, and other agents that support the Agent Skills format
github.com/voximplant/voximp…
Now live in our Grok Voice Agent client: @xai's new and improved grok-voice-think-fast-1.0 model. One-line change in your client config to use it.
Everything else in our connector stays the same. Give it a try!
voximplant.com/docs/referenc…
Shipping today: Voximplant Secrets. A dedicated credential store for VoxEngine scenarios — role-gated, CI/CD-ready, single-call retrieval.
A clean foundation for every third-party API our developers integrate with.
voximplant.com/blog/secrets-…
