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 / T2-H2
What this task measures
The caller is on the line with Kestrel Electronics for one locked request: Ohio drone restocking exemption. They keep repeating that ask through a callback offer, a different ticket, or an unrequested fix. We included it as a hard returns and refunds case because window and restocking math has to be spoken before a return is opened; state-law exclusions are the usual miss. The agent succeeds only if it stays on that request, reaches the outcome by following the verification then returns path, and leaves the record exactly as this case requires: the right row written or cleared, and nothing extra.
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 $0.248
0:00 / 2:49 · audio · trace 813249 · 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 final database state did not match. Combined cannot pass unless every gate does.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
+ from: reception
+ to: verification
Output
None recorded.
None recorded.
Parameters
+ full_name: Owen Tsai
+ order_number: KE-4487740
Output
+ ok: true
Parameters
card_last4: 4482
card_last4: 4482
postal_code: 43215
postal_code: 43215
Output
+ ok: true
Parameters
None recorded.
None recorded.
Output
+ ok: true
Parameters
+ from: verification
+ to: returns
Output
None recorded.
None recorded.
Parameters
order_number: KE-4487740
order_number: KE-4487740
Output
+ ok: true
Parameters
opened: true
opened: true
order_number: KE-4487740
order_number: KE-4487740
+ sku: SKU-DRN-4400
Output
+ ok: true
Parameters
opened: true
opened: true
order_number: KE-4487740
order_number: KE-4487740
+ reason: Unwanted item
+ sku: SKU-DRN-4400
Output
+ ok: true
Parameters
+ confirmation_token: KE-RTN-4417
+ fee_disclosed_acknowledged: true
Output
+ ok: true
Parameters
+ rma_number: RMA-7798555
Output
+ ok: true
Parameters
+ reason: Caller confirmed they are all set after starting return and sending label.
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Extra: create_return_label, 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_returns
Actual path
transfer_to_verification → transfer_to_returns
Rows that differ between the expected hangup and the actual hangup.
extra · holds · KE-RTN-4417
Expected
Actual
+ customer_id: cust_owen
+ kind: return
+ summary: Returning the Skyward Vireo 3 Drone, $999.99. There's no restocking fee, so the full $999.99 comes back. It goes back to the card ending 4482.
+ token: KE-RTN-4417
extra · return_labels · 1
Expected
Actual
+ id: 1
extra · rmas · KE-4487740
Expected
Actual
+ customer_id: cust_owen
+ order_number: KE-4487740
+ status: open
Locked caller script. This is the technical definition. Copy it as written.
Open with exactly: "I want to return the Skyward Vireo 3 drone I bought in Ohio. The box is opened." Give identity when asked. Order KE-4487740. If they ask why you are returning it, say exactly: "I changed my mind." and continue the return; do not wrap up on that turn. If they quote 15 percent, say it was purchased in Ohio. Confirm the return only after they say there is no restocking fee and the full $999.99 comes back. Do not ask for a label on this call. Once they have started the return and confirmed it, say exactly: "Thank you, that's all I needed." and end the call. Do not start a second request.
