All termsGlossary

What is voice latency?

Voice latency is the delay between a caller finishing what they say and hearing the system’s reply. It also covers delay in the audio itself as it travels across the network.

Why it matters

Conversation has a rhythm, and people are quick to notice when it breaks. A pause that callers notice makes a call feel unnatural. They start to wonder whether they were heard, repeat themselves or speak again just as the reply begins, and the two sides talk over each other. Low delay matters more on the phone than in chat, because silence on a call has no typing indicator.

How it works

The total delay is the sum of several steps. The system must decide the caller has finished, turn the speech into text, work out a reply, turn that reply into audio and send it back over the network. Each step adds time. Teams reduce it by streaming each stage so the next can start early, placing servers close to callers and keeping lookups in other systems fast.

Example

A team tests its voice agent and finds that replies feel slow when a caller asks about an order. They time each stage and see that most of the wait comes from the order lookup, not from speech processing. They have the agent say “Let me check that for you” while the lookup runs, and speed up the query. The pause no longer feels like dead air.

Get Started

Let’s build AI agents, together.

Talk with us to see the top three use cases for AI agents at your business, and how quickly you could go live.

Live in 7 days · Works inside the tools your team already uses · Hands off with full context