Roark Live Bench · Suite v1.3 · Sweep of 23 October 2026
Voice AI models, ranked on live phone calls.
Every speech-to-text, language model and text-to-speech model, run in every combination on the same real call to the same agent. A model is ranked on every stack it runs in; the stacks are the evidence.
Our picks
Task completionMedian responseCost per minute- Best overallAssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Cartesia Sonic-3.697%3.9 s$0.055
- Best valueAssemblyAI Universal-3.5 Pro + Google Gemini 3.8 Flash + Cartesia Sonic-3.597%4.3 s$0.048
- FastestOpenAI Whisper-1 + Google Gemini 3.8 Flash + OpenAI TTS-1 HD88%0.5 s$0.023
- Best speech-to-speechgrok-voice-think-fast-2.082%2.1 s$0.083
- Best in customer supportAssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + ElevenLabs v4 Turbo100%3.9 s$0.096
- Best in healthcareAssemblyAI Universal-3.5 Pro + Google Gemini 3.8 Flash + Cartesia Sonic-3.696%4.4 s$0.041
- Best in home servicesAssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Cartesia Sonic-3.596%4.0 s$0.058
- Best in insuranceAssemblyAI Universal-3.5 Pro + OpenAI GPT-4o + Cartesia Sonic-3.698%3.2 s$0.042
The read
- 430 of 642 cascades complete the task in at least 90% of calls, and the top 10 sit within 1 point of each other. What separates the top stacks is speed and cost.
- Median response time runs from 0.5 s to 6.0 s on the same agent and the same caller, a spread of 5.5 s that comes entirely from the models behind it.
- No stack wins everything. The fastest, OpenAI Whisper-1 + Google Gemini 3.8 Flash + OpenAI TTS-1 HD at 0.5 s, ranks 530 of 642 on task completion; the most complete is Speechmatics Linden-1 (own VAD) + OpenAI GPT-6.1 Sol + OpenAI TTS-1 HD at 99%.
- The language model matters most: swapping it moves task completion by 7 points, against 3 for the speech engine and 2 for the transcriber.
Cascade vs Speech-to-Speech
The first choice to make. A cascade chains three models: one transcribes, one decides what to say, one speaks. A speech-to-speech model does all three at once. Both run the same agent on the same calls here.
- Cascades lead on task completion, tool use and time to first word; speech-to-speech leads on response time and cost per task. Neither wins outright, so the right choice depends on what your calls need.
- The biggest difference is task completion: 95% for cascades against 77% for speech-to-speech.
| Measure | Cascade | Speech-to-speech |
|---|---|---|
| Task completion | 95% (leads) | 77% |
| Tool use | 72% (leads) | 55% |
| Time to first word | 0.7 s (leads) | 1.3 s |
| Response time | 3.6 s | 2.6 s (leads) |
| Cost per task | $0.15 | $0.15 (leads) |
The three strongest of each, averaged. Cascades: Speechmatics Linden-1 (own VAD) + OpenAI GPT-6.1 Sol + Cartesia Sonic-3.5; AssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Rime Coda; Speechmatics Linden-1 (own VAD) + OpenAI GPT-6.1 Sol + Rime Coda. Speech-to-speech: grok-voice-think-fast-2.0; gemini-3.8-live; amazon.nova-2-sonic-v1:0. Each measure is scored 0 to 100 on its own scale, further out is better; the table gives the same figures in their own units.
Which to use
- Choose speech-to-speech when the pause matters more than the outcome: short transactional calls, callers who interrupt, queues where every half second costs you callers. Expect to pay more, and choose the model carefully (gpt-live-1 (backend gpt-6.1-sol) leads here).
- Choose a cascade when the call has to complete, when cost matters, or when you need to see which stage to fix. It is 13 points ahead on task completion here.
- Costs are what each stack’s providers charged for these calls, at the rates in effect when they ran. Telephony and hosting are not included.
Speech-to-Text
The transcriber turns the caller's speech into text. It decides what the agent thinks it heard, and nothing after it can start until it finishes.
- AssemblyAI Universal-3.6 Pro, AssemblyAI Universal-3.5 Pro, Deepgram Nova-3 and Speechmatics Enhanced (realtime) are level: none leads by more than the margin of error. Of them, AssemblyAI Universal-3.6 Pro costs least, at $0.030 a call. Avoid Speechmatics Linden-1 (own VAD): last on semantic word error rate (2.2%).
- Speechmatics Linden-1 (own VAD) delivers its transcript in 0.49 s, OpenAI Whisper-1 in 1.93 s: 1.44 s added to every turn of the call.
- AssemblyAI Universal-3.6 Pro is the most accurate: it changes the meaning of 0.9% of the caller's words, against 2.2% for Speechmatics Linden-1 (own VAD).
- In 341 of 648 cascades the transcriber is the slowest step of the turn, whichever model runs it.
Open a model to see where it ranks and the stacks it works best in.
Language model
The language model decides what the agent says and does. It has the most influence on whether the task gets done, and usually takes the largest share of each turn.
- Pick OpenAI GPT-6.1 Sol: it leads on task completion (95%), tool use (71%) and instruction following (4.2 / 5). If speed matters more, Anthropic Claude Haiku 4.5 is the quickest, at 0.36 s, and completes 91%. Avoid OpenAI GPT-5.1: last on tool use (33%) and instruction following (4.1 / 5).
- Anthropic Claude Haiku 4.5 starts answering in 0.36 s, Google Gemini 3.8 Flash in 1.47 s: 1.11 s added to every turn of the call.
- OpenAI GPT-6.1 Sol uses its tools most reliably: it handles its tools without a mistake on 71% of calls, against 33% for OpenAI GPT-5.1.
- In 161 of 648 cascades the language model is the slowest step of the turn, whichever model runs it.
Open a model to see where it ranks and the stacks it works best in.
Text-to-Speech
The speech engine is the voice the caller hears, and the last step before they hear anything.
- Cartesia Sonic-3.6, Cartesia Sonic-3.5 and ElevenLabs v4 Turbo are level: none leads by more than the margin of error. Of them, Cartesia Sonic-3.5 costs least, at $0.165 a call. Avoid OpenAI TTS-1 HD: last on readback accuracy (80%), word error rate (8.7%) and time to first audio byte (2.09 s).
- ElevenLabs v4 Turbo starts speaking in 0.11 s, OpenAI TTS-1 HD in 2.09 s: 1.97 s added to every turn of the call.
- Cartesia Sonic-3.6 is the most accurate: it reads codes, amounts and dates back correctly 87% of the time, against 80% for OpenAI TTS-1 HD.
- Cartesia Sonic-3.6 sounds the most human: 3.8 / 5, against 3.4 / 5 for ElevenLabs Flash v2.5.
- In 99 of 648 cascades the speech engine is the slowest step of the turn, most often with OpenAI TTS-1 HD (54 of them).
Open a model to see where it ranks and the stacks it works best in.
Speech-to-Speech models
One model listens, reasons and speaks, so there is nothing to swap and no stages to split. Each is judged on the whole call, with the same calls and the same scoring as every cascade.
- The model matters as much as the architecture: gpt-live-1 (backend gpt-6.1-sol) completes 86% of tasks, ElevenLabs Realtime v1 Mini 42%.
- grok-voice-think-fast-2.0 answers in 2.1 s against 4.1 s for gemini-2.5-flash-native-audio-preview-12-2025.
- Price varies as much as quality: $0.06 per completed task for amazon.nova-2-sonic-v1:0, $0.69 for gpt-realtime-2.
Open a model for its full report, every figure it was scored on and the calls behind them.
Every stack
The evidence behind the model boards: each combination on its own calls, scored on the same recipe. Open a stack for its full report, every figure it was scored on and the calls behind them.
Each score carries its 95% margin of error. A rank is the best place a stack could hold once every margin is allowed for, so stacks the calls cannot tell apart share one and are listed by the low end of their margin; hover a rank for the worst place it could hold. 92 stacks are level with the leader: their calls cannot tell them apart. Among them, 31 Google Gemini 3.8 Flash, 29 OpenAI GPT-6.1 Sol, 19 Anthropic Claude Sonnet 5.5, 9 Anthropic Claude Haiku 4.5, 2 OpenAI GPT-4o mini, 1 OpenAI GPT-4.1 and 1 speech-to-speech; 48 AssemblyAI, 23 Deepgram and 20 Speechmatics; 48 Cartesia, 33 ElevenLabs, 9 Rime and 1 OpenAI. A cascade's figures are 25% its own calls and 75% what every other stack shows about its transcriber, language model and voice.
Not ranked here for missing data: Speechmatics Linden-1 + OpenAI GPT-6.1 Sol + OpenAI TTS-1 HD (no response time or tool use or human-likeness or readback accuracy or voice consistency).
By industry
Each industry ranked on its own calls: the same five callers, its own tasks and tools. A stack has a handful of calls in each, so a cascade is ranked on what its transcriber, language model and voice show across every stack in that industry, and a speech-to-speech model on its own calls there. Pairings show on the main board. Every panel is on one scale, so a bar reads against the other industries as well as its own.
Healthcare
Reschedule an appointment. A patient moves a dermatology appointment around a work schedule, corrects their date of birth, narrows the times twice, and asks about the change fee, parking and insurance.
Home services
Urgent repair: triage and booking. A caller with a burst pipe needs safety advice first, then an emergency visit. The unit on file is out of date, the callback number gets corrected, and the caller pushes for an earlier visit.
Insurance
Report a new auto claim. A driver who was just rear-ended reports a claim, corrects the policy number, spells the other car’s plate, asks about the deductible, a rental and their premium, and needs a tow.
Customer support
Internet outage troubleshooting. A technical caller who has already rebooted the gateway pushes back on scripted steps, wants the line fixed, and asks for a bigger credit for the downtime.
How it works
Every number on this page comes from real phone calls, placed and scored the same way for every stack.
Real calls, four industries, five callers
Roark places real phone calls over the public phone network to a reference agent for each of four industries: rescheduling a clinic appointment, booking an urgent home repair, opening an auto insurance claim and fixing an internet outage. Each agent has its own prompt and tools, identical for every stack. The tasks carry the details real calls trip on: a corrected date of birth, a policy number spelled out, a constraint added halfway through.
Every industry is called by the same five simulated callers: a quiet US English line, a Spanish-accented caller on a city street, an Indian English caller in a busy café, a flustered Southern caller with a baby crying, and a fast caller who interrupts over construction noise. They are equally clear about what they want, so a call fails because of the agent, not the caller. Only the models change, so any difference on the board comes from them.
Each caller rings each industry once for a cascade and four times for a speech-to-speech model. A cascade's figures also draw on its parts, heard in dozens of other stacks; a speech-to-speech model has only its own calls, so it takes more of them.
That makes industries comparable over the same callers, and callers over the same tasks, and the board can be re-ranked by either. One caller within one industry is too few calls to judge, so the board never shows it.
655
stacks
20
calls per stack
13,876
calls scored
Every model, in every combination
9 transcribers, 8 language models and 9 voices are run in every combination with each other, beside 13 speech-to-speech models. A model's own figures pool every stack it runs in, adjusted for its partners in each one, so every model is compared as if it ran with the same partners. A transcriber's figure rests on hundreds of calls with every language model and every voice, not on one lucky pairing.
That measures a model where it actually works: inside a running agent, on live phone audio, through interruptions, corrections and real turn-taking. A test that feeds a model clean audio files or one prompt at a time never sees any of that.
One score, weighted by what matters on a call
Each section ranks on a Roark score. Every figure scores 0 to 100 between two published points: full marks where doing better makes no difference to the caller, zero where the call breaks down. Each then counts at its weight, and the weights follow one order everywhere: whether the caller got what they rang for, then whether the agent got it right, then speed.
A stack, cascade or speech-to-speech, is scored on the outcome and the reply first, then on what each of its components got right on its own calls: the words its transcriber got wrong in a way that changes the meaning, whether its voice read codes and amounts back right, how human it sounded, and whether its tools were handled. A component that leads its section lifts every stack that carries it, by the same recipe for every stack. A figure a stack cannot carry, such as a semantic word error rate for a speech-to-speech model with no transcript, is credited at the field median, so the stack is neither flattered nor punished for it.
On a full sweep each cascade's figure is 25% what its own calls measured and 75% what every other stack shows about its transcriber, language model and voice, fitted across the whole sweep. The split is the same for every stack, so a lucky run does not lead and a stack's score does not move with how noisy its calls were. In an industry a stack has a handful of calls, so a cascade's figure there is what its transcriber, language model and voice show across every stack in that industry's calls, and a speech-to-speech model's is its own calls there. How a particular pairing beats or misses its parts shows on the main board, where each stack has all of its calls. The best overall pick, on the main board and in each industry, is the stack whose score is highest at the low end of its margin of error, so a few lucky calls cannot carry a stack past one whose score rests on hundreds. Stacks that share a rank are listed the same way, so the first row is the pick.
That order comes from real-world voice agent calls, where getting the conversation right predicts how a call ends more strongly than speed does. The latency points come from the same calls: replying faster than about 2 seconds made no measurable difference, and calls went measurably worse once the agent took more than about 3 seconds to reply.
Stacks and speech-to-speech
| Figure | Weight | Full marks | Zero |
|---|---|---|---|
Task completionscenario_adherence_check | 21% | 100% | 80% |
| An agent that completes fewer than four calls in five is not fit to deploy. On a full sweep, a cascade's figure is 25% what its own calls (20 or more) measured and 75% what every other stack shows about its transcriber, language model and voice, the same split for every stack, so a lucky run does not lead. Where the sweep is too small for that, and for a speech-to-speech model, the figure is the stack's own. In an industry a stack has a handful of calls, so a cascade's figure there is what its transcriber, language model and voice show across every stack in that industry's calls, and a speech-to-speech model's is its own calls there. How a particular pairing beats or misses its parts shows on the main board, where each stack has all of its calls. | |||
Response timeresponse_time | 17% | 2.0 s | 6.0 s |
| In real calls, replying faster than 2 s made no measurable difference, and calls went worse past 3 s. | |||
Instruction followinginstruction_follow | 13% | 5.0 / 5 | 3.0 / 5 |
| Every model scores between 3 and 5 on the judge's 1 to 5 scale, so that is the range. | |||
Semantic word error ratestt_semantic_word_error_rate | 13% | 0.5% | 3% |
| At 0.5% the best transcribers change the meaning of one word in two hundred on phone audio; at 3% one word in thirty sends the agent the wrong way. | |||
Tool usetool_use | 13% | 100% | 50% |
| An agent that mishandles its tools on half its calls cannot be trusted with the systems it runs on. | |||
Human-likenessvoice_human_likeness | 8% | 4.5 / 5 | 2.5 / 5 |
| 4.5 out of 5 is close to a person on the phone; 2.5 sounds plainly synthetic. | |||
Time to first wordtime_to_first_word | 4% | 1.0 s | 4.0 s |
| A greeting within 1 s sounds natural; 4 s of silence sounds like a dead line. | |||
Readback accuracytts_entity_readback_accuracy | 4% | 100% | 60% |
| Reading a code, amount or date wrong breaks the task it was for; the weakest voices get two in five wrong. | |||
Word error ratetts_word_error_rate | 4% | 2% | 12% |
| At 2% the voice says what it was given; at 12% one word in eight comes out wrong. | |||
Voice consistencyvoice_consistency_check | 4% | 100% | 80% |
| A voice that changes speaker on one call in five is not one a business can put its name to. | |||
Shown, not scored: Turn budget, Barge-in handlingbarge_in_yield_appropriateness
Counted only with price factored in: Cost per task
Language model
| Figure | Weight | Full marks | Zero |
|---|---|---|---|
Task completionscenario_adherence_check | 36% | 100% | 80% |
| An agent that completes fewer than four calls in five is not fit to deploy. | |||
Tool usetool_use | 27% | 100% | 50% |
| An agent that mishandles its tools on half its calls cannot be trusted with the systems it runs on. | |||
Instruction followinginstruction_follow | 18% | 5.0 / 5 | 3.0 / 5 |
| Every model scores between 3 and 5 on the judge's 1 to 5 scale, so that is the range. | |||
Time to first tokenllm_time_to_first_token | 18% | 0.50 s | 2.00 s |
| Under 0.5 s is not noticeable; at 2 s the first token alone takes two thirds of the 3 s past which real calls go worse. | |||
Shown, not scored: Prompt cache hitsllm_prompt_cache_hit_rate
Counted only with price factored in: Cost per call
Speech-to-Text
| Figure | Weight | Full marks | Zero |
|---|---|---|---|
Semantic word error ratestt_semantic_word_error_rate | 71% | 0.5% | 3% |
| At 0.5% the best transcribers change the meaning of one word in two hundred on phone audio; at 3% one word in thirty sends the agent the wrong way. | |||
Time to final transcriptstt_time_to_final_transcript | 29% | 1.00 s | 2.50 s |
| The fastest transcribers finish in about 1 s; past 2.5 s the transcript alone uses most of a natural reply time. | |||
Shown, not scored: Word error ratestt_word_error_rate
Counted only with price factored in: Cost per call
Text-to-Speech
| Figure | Weight | Full marks | Zero |
|---|---|---|---|
Readback accuracytts_entity_readback_accuracy | 32% | 100% | 60% |
| Reading a code, amount or date wrong breaks the task it was for; the weakest voices get two in five wrong. | |||
Word error ratetts_word_error_rate | 21% | 2% | 12% |
| At 2% the voice says what it was given; at 12% one word in eight comes out wrong. | |||
Time to first audio bytetts_time_to_first_byte | 21% | 0.20 s | 1.50 s |
| Under 0.2 s is not noticeable; 1.5 s is half the point where calls go worse. | |||
Human-likenessvoice_human_likeness | 16% | 4.5 / 5 | 2.5 / 5 |
| 4.5 out of 5 is close to a person on the phone; 2.5 sounds plainly synthetic. | |||
Voice consistencyvoice_consistency_check | 11% | 100% | 80% |
| A voice that changes speaker on one call in five is not one a business can put its name to. | |||
Counted only with price factored in: Cost per call
Factor in price and cost takes 25% of the score; the other weights keep their proportions. Price is scored on a log scale, so each doubling costs the same points.
A figure listed as shown, not scored is on the section’s boards and outside its score. Where a section has not collected a scored figure yet, the others share its weight.
Every call recorded and judged
Every call is recorded. Judge models score each transcript for the outcome and the conversation, and the timings come from the recording. Every figure links to the calls behind it.
Roark is on both sides of every call: it plays the caller, so it keeps the exact text its caller said, and it runs the agent under test, so it sees what the agent’s transcriber heard and what its model wrote for its voice. The transcript and voice figures come from lining those up, with no hand-made transcript in between. Each carries a margin from the synthetic voice or Roark’s transcriber in the loop. Every model faces the same ones under the same conditions, so the figures compare models on the same calls through the same voice and the same ear, though those slips need not fall evenly on each model. The rates will not match a vendor’s figure on a curated set; read them against each other.
An example: one call from this sweep
Speechmatics Linden-1 (own VAD) + OpenAI GPT-6.1 Sol + OpenAI TTS-1 HD, 23 October 2026. Highlighted turns are the ones that affected a score. Every stack's report links to its calls.
See how your own agent compares
Roark runs the same kind of calls against your agent, on your scenarios, with the same scoring.
Rank on Roark score
#294of 654 stacks
47 / 100
Leader AssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Cartesia Sonic-3.6, 68 / 100
Where it ranks
Task completion
Task completion
How much of the caller’s task the agent completed: the share of the scenario’s checks a judge model found met in the transcript, averaged over calls. Before v1.3, the share of calls judged 4 or better out of 5.
#657 of 661 on the board · best 99%
Flow adherence
How closely the agent followed the intended conversation flow, judged per call. 5 = fully followed.
#100 of 661 on the board · best 5.0 / 5
Instruction following
How well the agent stuck to the instructions in its own prompt. 5 = followed throughout.
#12 of 661 on the board · best 5.0 / 5
Turn-taking
Barge-in handling
When the caller interrupts, how often the agent made the right call: yielding to a real correction, and holding the floor through a filler.
#5 of 6 on the board · best 100%
Incorrect interruptions
How often the agent interrupted the caller when it should have kept listening.
#1 of 661 on the board · best 0%
Overtalk
Share of the conversation where both sides were speaking over each other.
#577 of 661 on the board · best 0.0%
Agent self-cutoffs
Times the agent cut its own sentence off mid-word.
#1 of 661 on the board · best 0.0 per call
Understanding & effort
Caller effort
How hard the caller had to work: repeating themselves, correcting, re-explaining. 1 = effortless.
Not scored by this sweep.
Repeated-info loops
Times per call the agent re-asked for something the caller had already given.
#180 of 493 on the board · best 0.0 per call
Voice
Human-likeness
How human the agent's voice sounds, scored by an audio model on intonation, rhythm and expressiveness. 5 = hard to tell from a person.
#78 of 659 on the board · best 4.3 / 5
Voice consistency
Share of calls where every line the agent spoke was the same speaker, with no change of speaker, gender or language voice partway. It checks who is speaking, not how they sound.
Not scored by this sweep.
#1 of 659 on the board · best 100%
Latency
Response time
The silence between the caller finishing and the agent starting to speak, measured from the recording on every turn of the call.
#53 of 660 on the board · best 0.50 s
Time to first word
How long the caller waits before the agent first speaks after pickup.
#372 of 661 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#55 of 661 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 73% | 3.02 s | 76% |
| Home services | 74% | 2.29 s | 80% |
| Insurance | 61% | 2.43 s | 88% |
| Customer support | 72% | 2.72 s | 91% |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 75% | 2.70 s | — |
| Spanish-accented English, city street | 65% | 3.02 s | — |
| Indian English, busy café | 64% | 3.15 s | — |
| Southern US, flustered, crying baby | 70% | 3.02 s | — |
| Fast and interrupting, construction outside | 77% | 0.04 s | 82% |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
AGENTOh wow, that sounds really stressful, especially when you’re working from home. A blinking orange light can definitely mean something’s off, and with a road crew nearby, it could be related to a line issue. You’ve already rebooted the gateway and checked cables—great troubleshooting. You could try plugging in directly to the modem if you have one, and then check your provider’s outage map or contact support to confirm if row digging affected the fiber or line. If you’d like, I can walk you through what to say when you call, like what details to mention so they can resolve it faster.
CUSTOMERYeah, my internet has been down for about two hours now. I work from home, so it's been pretty disruptive. There's a road crew digging up the street outside, and I'm wondering if that might be the cause. I already rebooted the gateway twice and checked every cable. The internet light is blinking orange. Can you help me get this sorted out?
AGENTOkay, let me pull up your account and check for any outage information.
CUSTOMERThanks, but I'm already on the call with you. I've given all the details: the blinking orange light, the road crew, and that I already rebooted and checked cables twice. My account number is BW883017. Sorry, last digit is 1. Can you check if there's an outage in my area or if the road work caused this?
AGENTAlright, I checked your account BW 8 8 0 3 1 7 1, and there’s no known outage or maintenance in your area.
Next, I ran a line test. The signal to the gateway looks good, but the gateway itself wasn’t registered.
So I sent a remote reset, and now the gateway is registered and online. Since it may take about two minutes for your devices to reconnect, you might see the internet come back shortly.
If it doesn’t, I can help schedule a technician visit.
CUSTOMEROkay, the internet light just went solid white. How long should it take for my devices to get back online? And since I lost two hours of work, can I get a credit for the downtime?
AGENTGot it, let me think through how your devices should reconnect and how the credit works.Nice, a solid white light is a good sign. It usually means the gateway is up and running, and it can take about two minutes for your devices to reconnect. If anything still isn’t coming back online after that, we can try resetting or scheduling a technician.
As for the downtime credit, I applied a ten dollar service credit for the outage time, and it will show up on your next bill. If you notice it not applying or anything feels off, I’m right here to help.
CUSTOMERI lost two hours of work. Can you make it more than that?
AGENTI understand it’s really frustrating to lose work time, and I hear you. A service credit of $10 has already been applied to your account, and it’ll appear on your next bill. For a larger credit, that’s something our billing team has to review based on the outage and your usage. If you’d like, I can help you explain the situation clearly when you contact billing.
AGENTYou’re very welcome! I’m glad I could help. Since the gateway’s back online andGoodbye! Take care, and
CUSTOMEROkay, I'll reach out to billing about that. Thanks for your help, Alex. Goodbye. Goodbye.
Healthcare
AGENT"Riverbend Health Clinic, this is Alex. How can I help you today?"
CUSTOMERYeah, I'm Nate Ross. I need to reschedule my dermatology appointment that's set for next Wednesday morning. I just got put on a work shift for that time, so I can't make it. Can we move it to something after 3pm instead? I'm on my lunch break, so if we can do this fast, that'd be great. 홍길동.
AGENTOkay, let’s get you rescheduled, quick and clean. Before I can change your appointment, I need to verify your record. What’s your full name and date of birth? Just the date in year-month-day format is fine. Once you share that, I’ll locate the dermatology appointment and look for slots after 3 p.m.
CUSTOMERGot it! Ja.
AGENTGreat, thank you. To verify your record, I need your full name and date of birth. You’ve given me your name, Nate Ross. What’s your date of birth in year-month-day format? With that, I can look up your existing appointment and find a slot after 3pm that works.Perfect—Nate, I have your full name. I still need one more detail to verify your record: your date of birth, year-month-day format is perfect. Once you share that, I can look up your appointment and find available times after 3 p.m.
CUSTOMEROkay.
AGENTThanks, Nate. I just need your date of birth. You can say it as year, month, day. Once I have that, I’ll look up your appointment and see what opens after 3 p.m.
CUSTOMERMarch 17, 1948. Sorry, no, 1984.
AGENTThanks for the update—let me just quickly confirm your details and check what’s available.
AGENTGot it, thanks for clarifying your
CUSTOMERMike, go ahead. Okay.
AGENTThanks, Nate. I confirmed your record and found your current appointment. It’s a dermatology appointment on next Wednesday at 9:15 with Dr. Okafor.
For after 3 p.m., the options available are Tuesday at 345 p.m. and Thursday at 4:10 p.m.
Just let me know which one you prefer, or if you’d like to try a different day or time, and I’ll move it.
CUSTOMERRight.
AGENTI found your current appointment: dermatology with Dr. Okafor, next Wednesday at 9:15 a.m.
After 3 p.m., the available options are Tuesday at 345 p.m. and Thursday at 4:10 p.m.
Do you want to reschedule to Tuesday at 345 p.m., or Thursday at 4:10 p.m.? Let me know which one you prefer, and I’ll update it right away.
CUSTOMERTuesdays won't work. I pick up my daughter then. What about Friday?
AGENTLet me check Friday availability for Dr. Okafor so we’re not guessing.
AGENTGot it, Friday doesn’t have an after-3 p.m. slot. The only available time after 3 p.m. for your provider is Thursday at 4:10 p.m. Would you like me to reschedule your appointment to Thursday at 4:10 p.m.?
CUSTOMERYep, go ahead and move it to Thursday at 4:10 PM. That works!
AGENTAlright, let me update that appointment time for you.
AGENTAll set, Nate! Your appointment was successfully rescheduled to Thursday at 4:10 p.m. Your new confirmation code is R V B dash 3 M 8 P 2. Please plan to arrive 15 minutes early with a photo ID and your insurance card. Anything else you want to adjust or ask about?
CUSTOMERQuick question: is there a fee for changing the appointment? Alright. Okay.
AGENTLet me check what the clinic policy says about changing appointment fees.
AGENTI’m checking that policy, but it looks like the policy lookup is still running. If it takes longer, that’s just the delay. If you’d like, I can try again and confirm for you.
AGENTYes, Nate. Changes made more than 24 hours before your appointment are free. Changes later than that, or too close to the time, have a 40 dollar fee. Since we moved it to Thursday at 4:10 p.m., the fee would depend on how far out it was when you changed it.
CUSTOMERAlso, where should I park? Is there a specific lot for patients?
AGENTYes, patients can park for free in Lot C, with the entrance on River Road. If you’re arriving early, allow a few extra minutes to find the lot and get to the building.
CUSTOMERAll right, last thing, does Dr. Okafor accept Aetna insurance?
AGENTGood question. I don’t have access to Dr. Okafor’s insurance participation list from here, so I can’t confirm coverage for Aetna. The best option is to call the clinic or your insurance company and ask whether Dr. Okafor is in-network or accepts Aetna for dermatology. That’s the safest way to get accurate information.
CUSTOMERAll right, I'll check with them directly. Thanks for your help. Goodbye.
AGENTYou're welcome, Nate! Glad I could help. If you need anything else, just give us a call. Goodbye!
Home services
AGENT"Copperline Home Services, this is Alex. What can I help you with?"
CUSTOMERHi, there's a pipe under my kitchen sink that just burst. Water is spreading across the floor, and it's heading right toward a power strip. I need someone out here as fast as possible!
AGENTOh wow, that sounds really urgent. First, turn off the main water supply if you can, and unplug anything plugged into the power strip right away. If water is anywhere near an outlet or any electronics, get out of the area and call emergency services or the gas/electric company if it’s safe to do so. Then, call an emergency plumber. They’ll get someone out fast to fix the pipe and help prevent more damage.
CUSTOMERJa. Okay.
AGENTI’m here with you. This is urgent, so first things first: if you can, shut off the main water supply right away. Then keep yourself and anyone else away from the power strip and outlets near the water.
To get help out as fast as possible, I’ll need your service address, your unit number if there is one, a gate code or access instructions, and a callback number. Once I have those, I’ll look up your customer details, create an emergency service request, and check the earliest available technicians and fees so we can book quickly.
AGENTAlright, thank you. That’s helpful. Now, for safety, stay away from the water and the outlets near the floor. If the power strip is wet, don’t try to unplug anything right now—stay clear of it, and keep others away too.
Next, I’ll need a bit of information so I can find your account and book the soonest emergency visit. What’s the service address, including the unit if there is one? What’s the gate code or any access instructions? And what’s the best callback phone number?
If anyone is in immediate danger, like someone getting shocked, please call 911 or evacuate immediately.
Once I have your address and callback number, I’ll look up your customer details, create an emergency service request, and find the earliest available technician window. I’ll also tell you the service fee for your situation before booking, and I’ll need your clear agreement on that fee.
CUSTOMERMy address is 1480 Larkspur Avenue, unit 4D. I moved from 4B to 4D last month, so make sure it's for 4D. The gate code is pound two seven one nine. Call back number is 503-555-0139.
AGENTGot it, let me look up your customer info and check availability and fees before anything is booked.
CUSTOMERNo, 0193. Can you check the cost of the visit before I agree to anything? Aha! Okay, go ahead. We really need someone out here as soon as possible.
AGENTGot it, I’ll set up an emergency request, confirm the fee, and book the earliest available window.
CUSTOMERRight, please let me know who's coming and how I'll know they're on the way. Should I need the ticket number for this?
AGENTAll set,Alright, quick update. The technician booked is Dana. They’ll arrive between 2:00 PM and 4:00 PM today. For the fee, as a Copperline Care customer, the emergency callout is $99. Before anything is officially booked, that’s the amount being applied, so please confirm you’re okay with it. This is ticket number CPL-58210. The service address is 1480 Larkspur Avenue, Unit 4D, the gate code is pound 2 7 1 9, and the callback number is 503-555-0193. Dana will text your callback number about 30 minutes before arrival. If you’re comfortable with the $99 emergency fee, say “I agree.”
CUSTOMERGot it うん。 Got it!
AGENTPerfect, thanks for confirming.
Here’s the booking summaryService address is 1480 Larkspur Avenue, Unit 4D. Gate code is pound 2 7 1 9.Great, thanks for confirming. Just to recap for your safety: please stay away from the flooded area and any outlets that could be wet, and keep the power strip blocked off.
Here’s what I have on filethe service address is 1480 Larkspur Avenue, Unit 4D. The gate code is #2719. The callback number is 503-555-0192. Your ticket number is CPL-58210. The visit is booked for the 2:00 PM to 4:00 PM emergency window. The fee for your membership is $99, and you already agreed to proceed, which is great. The technician will text you about 30 minutes before arrival.
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
AGENTYes, I'm safe. I'm just...
CUSTOMERYes, I'm safe. Standing next
AGENTGot it, let me think
AGENTStanding next to you.
CUSTOMERはい。 I think you meant standing next to your car.
AGENTOkay, let’s continue. Quick heads-up: before we can open anything, I need to verify your policy. Can you tell me your policy number and the policyholder’s last name? Then we’ll go from there.
AGENTThanks for the correction—let me confirm your policy details and then we’ll move on.
CUSTOMERSure. The policy number is NG4471903. No, sorry, last one is 2. The last name is Ramirez. Okay, go ahead.
AGENTGreat, thanks. I’ve verified the policy. Just to confirm, the policy number is N G dash 4471903, and the last name on the policy is Ramirez. Now, can you tell me when the accident happened, where it happened, and what happened—just a quick summary. Also, any other driver information you have, like their name, insurer, and plate, and whether the police came. If your car can’t be driven safely, we can also set up a tow.
CUSTOMER리조트 About 20 minutes ago, I was stopped at a red light at SE 5th Avenue and Alder Street, and another car hit the back of my car. The police did not come. The other driver is Kevin Lung, insured with Pacific Shore. His plate is 7KXR214. The back bumper is crushed, and the rear wheel rubs, so I can't drive it. I do need a tow truck.
AGENTThanks for the details—let me get the claim started and arrange that tow.
AGENTOkay, here’s what I have. The claim number is NGM dash 904417. The other driver is Kevin Loong, insured with Pacific Shore. The license plate is 7Alright, your policy number is N G dash 4471903. The other driver’s plate is 7 K X R 2 1 4. Your claim number is N G M dash 904417. I’ve opened the claim, and a tow has been sent. The ETA is about 45 minutes. Your adjuster will be in touch within one business day.
CUSTOMEROk ろ。
AGENTGreat, thanks for confirming. I’ll keep the claim moving and the adjuster will follow up. If anything changes—like new details about the damage, the other driver, or if you need anything else—just let me know and we can update things.
CUSTOMEROkay, before I go, can you tell me how much I have to pay? Is there a deductible? Because he hit me, so I should not have to pay that, right? And do I get a rental car? Also, will my premium go up? Where will the car be taken? And could you repeat the claim number once more?
AGENTGood questions. Here’s what I can confirm from your coverage: your collision deductible is $500, and rental coverage is $40 per day for up to 30 days, starting when the claim is opened. Roadside and towing are covered up to 25 miles.
About who pays, those details can depend on the claim, liability, and your coverage. I can’t confirm whether your premium will increase based on this.
For the tow, it was dispatched, and it’s going to Eastside Towing, heading to the Northgate partner repair shop at 9th and Morrison.
Your claim number is N G M dash 9 0 4 4 1 7.
AGENTYes, I can restate that clearly.
About the deductibleeven if another driver hit you, it can still depend on coverage, fault, and how the claim is handled. From what was confirmed, your collision deductible is $500, and rental is $40 per day for up to 30 days. If the damage is covered, that rental should be available from the time the claim opens.
For the tow, Eastside Towing is going to the Northgate partner repair shop at 9th and Morrison, and the claim number is N G M dash 9 0 4 4 1 7. If you want, you can also follow up with the adjuster for any coverage questions and next steps.
CUSTOMEROkay, got it. So just to be clear, if he hit me I should not have to pay the deductible, right? And $40 per day for the rental car, up to 30 days, that's covered? And the tow is going to East Side Towing, then to the Northgate partner shop at 9th and Morrison? And the claim number is NGM-904417? Yes, that's right.
AGENTYou’re very welcome. I’m glad I could help. If anything else comes up, you know where to find me. Take care,
CUSTOMEROkay, that’s clear. Thank you for your help. Goodbye