Back to GlossaryTelephony
Latency
The time between when a customer finishes speaking and when the agent starts responding.
Latency is a critical metric in voice agents. High latency makes conversations feel unnatural and frustrating. Optimal latency is under 500 milliseconds — similar to natural human response time in conversation. Factors affecting latency: server location, ASR speed, LLM processing time, and TTS speed. According to Google Research, latency above 2 seconds reduces satisfaction by 40% (Source: Google Cloud, 2024). Yappr uses local servers in Israel for minimal latency.