Skip to main content
Back to challenges
Challenge

Build a Voice-Enabled Digital Trust Concierge with OpenAI Agents SDK and LiveKit

Create a real-time voice concierge that guides enterprises through Singapore's digital transformation blueprint. Using OpenAI Agents SDK for conversational orchestration and LiveKit WebRTC for low-latency voice interaction, your system will conduct interactive compliance readiness assessments efficiently.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build an ultra-low latency voice agent using OpenAI Agents SDK and LiveKit to assist organizations with Singapore digital transformation compliance.

Delivery guide

How work is evaluated

Evaluation

Test voice session turn completion and verification of assessment scoring.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

valid_maturity_score

Score increments correctly within allowed boundaries

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

response_latency_ms

Voice frame to voice response delay • target: 450 • range: 0-1000

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Build real-time WebRTC audio pipelines using LiveKit Agents framework

  • Integrate OpenAI Agents SDK for context-aware multi-turn conversations

  • Design structured checklist evaluation loops for enterprise tech adaptation

  • Ensure sub-500ms voice response latency over WebRTC streams

Resources and assets

Reference links and supporting material

Dataset notes

Transcribed enterprise technology readiness dialogues mapped to Singapore Digital Transformation Guidelines.

Challenge setup

Test voice session turn completion and verification of assessment scoring.

Challenge input

JSON containing transcript_turn and current_score

OpenAI Agents SDK

Autonomous agent logic and function execution engine.

LiveKit

Ultra-low latency audio communication infrastructure.

Evaluated output

JSON with updated_score, next_question, and recommendation

Evaluation checks
  • Score increments correctly within allowed boundaries
  • Voice frame to voice response delay • target: 450 • range: 0-1000
Configured evaluation evidence
  • Benchmark: Voice_Agent_Latency_Bench
  • Response Latency Ms target: 450
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • LiveKit · Required
  • OpenAI · Optional
  • Aim · Optional

Evaluation contract

  • valid_maturity_score · Weight 1
  • response_latency_ms · Weight 1

Metrics

  • response_latency_ms · Target 450

Benchmarks

  • Voice_Agent_Latency_Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing transcript_turn and current_score

    Output: JSON with updated_score, next_question, and recommendation

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework
  • LiveKit · Real-time Voice WebRTC

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Voice-Enabled Digital Trust Concierge with OpenAI Agents SDK and LiveKit