Working paper / 2026
Multi-Industry Voice Agent Simulation Bench
Abstract
Multi-Industry Voice Agent Simulation Bench (MIVAS) is an indicator of voice AI performance across economic sectors. MIVAS is comprised of tasks and verifiers with granular rewards executed in production-grade multi-agent RL environments across industries where voice AI is being adopted the fastest. Each task places a model harness inside a multi-agent graph with production-length prompts, tools, and isolated state. Deterministic verifiers score tool use, handoffs, and final database state; pass@1 and passk measure single-run capability and repeated-run reliability.
Clone the repository, pair a model harness with an industry task suite, and reproduce the evaluation environment.
Customer Support / T5-H3
What this task measures
The caller is on the line with Kestrel Electronics for one locked request: Second price match on the same soundbar. They keep repeating that ask through a callback offer, a different ticket, or an unrequested fix. We included it as a hard price match case because qualified-competitor math and the exclusion list are a policy maze that has to be spoken from the published rules. The agent succeeds only if it stays on that request, reaches the outcome by following the verification then orders path, and leaves the record untouched: no appointment, ticket, or note the caller never asked for.
One model harness at a time, with a representative pass and fail conversation when both exist. OpenAI Realtime 2.1 is the default view.
Recording
Conversation cost $0.381
0:00 / 4:14 · audio · trace 813437 · click a span to play, drag to scrub, scroll to pan
Audio
Speech Audio
Agent Audio
Handoffs
Tool Calls
Errors
Transcript
How a conversation passes
Combined pass is the logical and of three deterministic gates. Tools, handoff path, and final database state must all pass on the same conversation. One miss fails the call.
Combined
Fail
This sample fails because tools and final database state did not match. Combined cannot pass unless every gate does.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
+ topic: price match
Output
+ ok: true
Parameters
+ from: reception
+ to: verification
Output
None recorded.
None recorded.
Parameters
None recorded.
None recorded.
Output
+ ok: true
Parameters
+ full_name: Felix Moreau
+ order_number: KE4495108
Output
+ ok: true
Parameters
card_last4: 3390
card_last4: 3390
postal_code: 94612
postal_code: 94612
Output
+ ok: true
Parameters
None recorded.
None recorded.
Output
+ ok: true
Parameters
+ from: verification
+ to: orders
Output
None recorded.
None recorded.
Parameters
- order_number: KE-4495108
+ order_number: KE4495108
Output
+ ok: true
Parameters
competitor: Rivertide
competitor: Rivertide
competitor_price: 479.99
competitor_price: 479.99
+ in_stock: true
- order_number: KE-4495108
+ order_number: KE4495108
+ sku: SKU-AUD-7720
Output
+ ok: true
Parameters
+ confirmation_token: KE-PM-2286
Output
+ ok: true
Parameters
- competitor: Halcyon Mart
- competitor_price: 479.99
- order_number: KE-4495108
Output
None recorded.
None recorded.
Parameters
+ reason: Price match applied and caller confirmed no further needs.
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Parameter miss: get_order, quote_price_match. Missing: quote_price_match. Extra: get_policy, identify_customer, end_call.
Full industry architecture, same graph as the industry page. The left diagram marks the expected specialist path in green. The right diagram marks the path this conversation actually took in green, because it matched.
Expected path
transfer_to_verification → transfer_to_orders
Actual path
transfer_to_verification → transfer_to_orders
Rows that differ between the expected hangup and the actual hangup.
extra · holds · KE-PM-2286
Expected
Actual
+ customer_id: cust_felix
+ kind: price_match
+ summary: Aurelian Halo Soundbar was $549.99 and Rivertide has it at $479.99. That's $70.00 back to the card ending 3390.
+ token: KE-PM-2286
extra · price_matches · 1
Expected
Actual
+ customer_id: cust_felix
+ id: 1
+ order_number: KE-4495108
+ status: approved
Locked caller script. This is the technical definition. Copy it as written.
Open with exactly: "Match the Halo soundbar at Rivertide for seventy dollars less, then also match Halcyon Mart at the same price." Give identity when asked. Order KE-4495108, Halo. Confirm the first $70 Rivertide match. Then ask them to match Halcyon Mart too. Stay for the one-match-per-item refusal. Do not argue. Once they have finished what you asked and confirmed it, say exactly: "Thank you, that's all I needed." and end the call. Do not start a second request.
