Hi all, we’re based in the UK and using Twilio with Vapi. We’re currently experiencing around 4-5 seconds of latency on our calls. I know the UK ↔ US roundtrip adds ~200–300 ms, but that still doesn’t explain the full 4 seconds of lag we’re seeing. Is anyone else in the UK/EU running into the same issue, and is there any roadmap for EU/UK-based endpoints or regional hosting to reduce this? Also, is there anything we can do on our side to reduce the latency? It’s becoming a real issue, and pinpointing the cause is hard as we’ve already tried loads of things.

To help reduce call latency, try pre-loading your TTS and STT models at server startup to avoid cold-start delays during calls. For example, in your Node.js server, you can initialize your TTS engine before handling requests:

js
// At server startup
await ttsModel.load();
// Then handle incoming requests as usual

This ensures your models are ready for real-time processing and can significantly cut down on response times during live calls. For more optimization tips, see the [Vapi voice pipeline configuration guide](https://docs.vapi.ai/customization/voice-pipeline-configuration). Source: - [Voice pipeline configuration](https://docs.vapi.ai/customization/voice-pipeline-configuration)

This has not directly answered my question.

@Will what models are you using ?

Hi @Solaris we have tried a bunch of different models, but the latency has not changed much. I am more concerned there may be an issue with the VAPI and Twilio setup as I have seen reported before.

Hey @Will were you able to figure it out? We're facing the same issue on Germany.

Same in Bulgaria, I guess Vapi doesn't care. What a platform

Thanks for reaching out. Could you please share a Call ID for one of the calls where you experienced the latency issue? Once we have that, our team will investigate what’s causing the bottleneck and work on finding a suitable solution.