Physical Turnaround Latency Breakdown

Turn-by-turn ear-to-mouth timing. Shows where acoustic turnaround delay originates across the voice pipeline.

VAD / End-of-Turn Speech-to-Text LLM TTFT CRM Tool RTT TTS TTFB
ElevenLabs Agents Calibrated
Proprietary Turn Model · Scribe v1 ASR · Gemini 3.1 Flash-Lite · Flash v2.5 TTS (Best Voices)
400ms
LiveKit Agents Calibrated
Silero VAD v5 · Deepgram Nova-3 · Gemini 3.1 Flash-Lite · Inworld TTS-2 Flash
425ms
Retell AI Managed Calibrated
Retell Endpointer · Deepgram Nova-3 · Gemini 3.1 Flash-Lite · ElevenLabs Flash v2.5
460ms
Hover over or tap any pipeline segment to inspect component latencies and bottlenecks.
Fastest Turnaround 400ms ElevenLabs Agents
Measured Tool RTT — Run the fidelity suite for a measured D1 tool RTT
Task Effectiveness 100% Score Resolution + Tool Accuracy
Policy Safety 100% Pass 5/5 canonical guardrails

Unit Cost Economics & Scale Calculator

Unbundled component cost comparison, measured cost per resolved call, and interactive volume modeling across platforms.

10,000 calls
5.0 min / call
Component LiveKit Cloud Inference ElevenLabs Agents Retell AI Managed
Platform & Orchestration $0.004 / min (LiveKit Cloud) $0.080 / min (Voice + Turn Engine) $0.050 / min (managed markup)
Speech-to-Text (STT) $0.005 / min (Deepgram Nova-3) Included (ElevenLabs Scribe v1) Included (Deepgram Nova-3)
LLM Intelligence (Apples-to-Apples) $0.003 / min (Gemini 3.1 Flash-Lite) $0.003 / min (Gemini 3.1 Flash-Lite) $0.003 / min (Gemini 3.1 Flash-Lite)
Text-to-Speech (TTS) & Voice Realism $0.040 / min (Cartesia / Inworld) Included (Flash v2.5 · #1 Realism) $0.035 / min (ElevenLabs Flash v2.5)
Database & CRM Tool Compute $0.002 / min (D1 Edge API) $0.002 / min (D1 Edge Webhook) $0.002 / min (External Edge D1)
Measured Effectiveness & Resolution 100% eff · 100% resolved 100% eff · 100% resolved 100% eff · 100% resolved
Measured Cost / Resolved Call $0.2700 / call $0.4150 / call $0.4500 / call
Effective Unit Cost $0.054 / min $0.083 / min $0.090 / min
Calculated Monthly Spend $2,700 / mo $4,150 / mo $4,500 / mo
Annual LiveKit Savings
$21,600 / year
LiveKit versus Retell AI at the selected 10,000 calls / mo and 5.0 min / call

Policy & Security Fidelity Scorecards

Empirical verification of non-negotiable business rules, tool accuracy, and task completion across all 5 canonical regression scenarios.

Scope Guardrail: Pool Heater Refusal — Transcript Viewer

Select any scenario to inspect the turn-by-turn dialogue, policy guardrail assertions, and tool execution payloads.

Recent Measured Telemetry Runs (D1 History)

0 recorded runs
Timestamp Architecture Scenario Source Turns Cost Policy Effectiveness
Click "Run Fidelity Suite" or complete a live voice call to record empirical runs into D1.

Interactive WebRTC Multi-Agent Voice Console

Live microphone testing and ear-to-mouth verification across all three architectures. Evaluates acoustic turnaround, speech recognition, and real-time CRM tool invocation.

🔒 Security Protected

Demo Voice Engine — Security Gate

WebRTC audio calls stream bidirectionally to live AI backends (Retell AI, LiveKit, and ElevenLabs Agents). To prevent unauthorized API consumption during public deployment, enter your developer security token or password to unlock audio calling.

Protected by DEMO_VOICE_TOKEN / SEED_TOKEN

Live CRM Tool Execution Benchmark

Directly measure mid-call tool execution latency across the Worker edge backend.

D1 Schedule Cache (Edge Native) Recommended
12ms (RTT 47ms)
Single-digit local D1 query execution with edge caching
Uncached Workiz CRM Path Legacy WAN
418ms (RTT 453ms)
Simulated Workiz multi-page API network delay
Inspect Raw Backend Response Payload
Ready to test. Click a measurement button above to dispatch a live CRM request.