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
#153of 654 stacks
54 / 100
Leader AssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Cartesia Sonic-3.6, 68 / 100
Where it ranks
Its models
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.
#253 of 661 on the board · best 99%
Flow adherence
How closely the agent followed the intended conversation flow, judged per call. 5 = fully followed.
#1 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.
Not scored by this sweep.
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.
#256 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
Speech-to-Text
Semantic word error rate
The caller's words the agent's transcriber got wrong in a way that changes the meaning, as a judge reads each difference.
#95 of 648 on the board · best 0.0%
Word error rate
The caller's words the agent's transcriber missed, changed or invented, against what the caller actually said.
#243 of 648 on the board · best 1.2%
Text-to-Speech
Readback accuracy
Share of the codes, amounts, dates and phone numbers the agent wrote that its voice read out correctly. One wrong digit is a miss.
#506 of 603 on the board · best 100%
Word error rate
The share of the words the agent wrote that its voice dropped or changed, read back from Roark’s transcription of the recording; the same transcriber hears every voice.
#376 of 648 on the board · best 1.9%
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.
#390 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.
#260 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.
#123 of 661 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#248 of 661 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 92% | 3.42 s | — |
| Home services | 90% | 3.41 s | — |
| Insurance | 95% | 3.23 s | — |
| Customer support | 95% | 3.71 s | — |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 94% | 3.48 s | — |
| Spanish-accented English, city street | 94% | 3.56 s | — |
| Indian English, busy café | 94% | 2.90 s | — |
| Southern US, flustered, crying baby | 94% | 3.57 s | — |
| Fast and interrupting, construction outside | 90% | 3.38 s | 100% |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
CUSTOMERHi, Alex. My Internet's been down for about two hours. I work from home, so it's pretty disruptive. There's a road crew digging up the street outside. Could that be the cause? The gateway's Internet light is blinking orange. I already rebooted it twice and checked all the cables. Can you check what's happening on your end? My account number is BW883017. Sorry. Last digit is one.
AGENTThanks for the correction — I’ve got your account as BW883011. Here’s what I’m seeing on my end:
- No area outage at your address. - Your line signal to the gateway looks good. - Your gateway was not registered at first, so I sent a remote reset. - After the reset, the gateway is now registered and online. That blinking orange light you saw lines up with the “not registered” status, which the road work could have briefly triggered if they disturbed a line, but right now everything from our side looks healthy. Next steps on your side
1. Give the gateway about 2–3 minutes to fully come back up. 2. Then test on a device
- First by going to any website in a browser. - If that fails, try running a quick speed test (e.g., fast.com) once it’s working to confirm stability. If you’re still offline after 3–4 minutes
- Check whether any device (phone, laptop) can get online via Wi‑Fi or Ethernet. - Tell me exactly what you see on the gateway lights now (power, Internet, Wi‑Fi). If things come back up and you notice drops or slowness afterward, tell me and I’ll run another check and, if needed, arrange a tech—especially given the construction outside.
CUSTOMERThe Internet light just went solid white. How long until my devices reconnect? And since I lost two hours of work, can I get a credit for the downtime?
AGENTSolid white Internet light means the connection is back and the gateway is online. For your devices: - Most should reconnect within 1–3 minutes on their own. - If something doesn’t: - Toggle Wi‑Fi off/on (or unplug/replug Ethernet). - If it still won’t connect, forget the network and rejoin it with the Wi‑Fi password. On the credit: - I’ve applied a $10.00 service credit to account BW883011 for the ~2 hours of downtime. - You’ll see it on your next bill. If you want, tell me which device you need online first (work laptop, VPN, VoIP phone, etc.), and I’ll walk you through getting that one reconnected and checked.
CUSTOMERI lost two hours of work. Can you make it more than that?
AGENTI can’t change the amount that was just applied, but I can add a note on your account about the impact (lost work time, you work from home) so that if you have any further issues this month, we can reference it when adjusting credits. Right now, the system approved a $10.00 credit on account BW883011, which will appear on your next bill. That’s the maximum I’m able to issue for this specific short outage. If the connection becomes unstable again—especially with that road work going on—contact us right away. We can document additional downtime and, if needed, add more credit based on the total impact. In the meantime, do you want to quickly confirm your work setup (laptop + VPN, calls, etc.) is running smoothly now?
CUSTOMERNo. That's fine. I'll keep an eye on it and reach out if there's another issue. Thanks for your help, Alex. Goodbye.
Healthcare
AGENTRiverbend Health Clinic, this is Alex. How can I help you today?
CUSTOMERHi, Alex. I'm calling to reschedule my dermatology appointment next Wednesday morning. I just got assigned a work shift then, so I need to move it. Can we do that now? I really don't have much time.
AGENTYes, we can reschedule it now. I’ll be as quick as possible. First, I need to verify your record. Please tell me:
1. Your full name (first and last), and 2. Your date of birth (YYYY-MM-DD).
CUSTOMERNate Brooks, 03/17/1948. Sorry. No. 1984. 03/17/1984.
AGENTThanks, Mr. Brooks. I’ve pulled up your dermatology visit with Dr. Okafor next Wednesday at 9:15 AM. You mentioned you only need to move it off that morning. Here are the soonest other options with Dr. Okafor:
- Tuesday at 900 AM - Tuesday at 3:45
CUSTOMERYeah.
AGENTGot
CUSTOMERYeah. Tuesdays don't work after all. Pick up my daughter then. Is there any Friday slot after 3PM?
AGENTI’m not seeing any Friday openings with Dr. Okafor after 3:00 PM. The only Friday slot showing right now is 11:30 AM. The other current options
CUSTOMERAlright. What about Thursday? Anything after 3PM?
AGENTYes—there is one Thursday option after 3 PM:
- Thursday at 410
CUSTOMERGreat.
CUSTOMERThat works. Go ahead and book Thursday at 04:10PM.
AGENTYou’re all set. I’ve moved your dermatology appointment with Dr. Okafor from next Wednesday at 9:15 AM to:
- Thursday at 410 PM, at Main Campus, Building B Your new confirmation code is RVB-3M8P2. Please arrive 15 minutes early with a photo ID and your insurance card. Anything else you need before you go?
CUSTOMERYeah.
CUSTOMERYeah. Quick question. Is there any fee for changing the appointment?
AGENTNo, there’s no fee for this change. Our policy is: changes
CUSTOMERYeah.
CUSTOMEROkay. Got it. Where should I park for the appointment?
AGENTFor For your appointment at Main Campus, Building B, you can park in our free patient parking:
- Lot C, entrance
CUSTOMERYep.
CUSTOMERI know it's in Building B. Just tell me where to park. I need to wrap this up.
AGENTUnderstood. Park in Lot C, patient parking, entrance on River Road.
CUSTOMERAlright. Last thing. Does doctor Okafor accept Aetna insurance?
AGENTI don’t have insurance-plan details for individual doctors in my system,
CUSTOMERMhmm.
AGENTI don’t have insurance
CUSTOMEROkay.
AGENTI don’t have access
CUSTOMERRight.
CUSTOMERThat's fine.
AGENTI don’t have access to which specific plans Dr. Okafor is in-network with, including Aetna. To be sure, you can:
-
CUSTOMERUh-huh.
AGENTI don’t have access to Dr. Okafor’s insurance panel, including Aetna, in this system. To confirm Aetna coverage as fast as possible, you have two options:
1) Call the clinic’s main line and ask billing to check Dr. Okafor against your exact Aetna plan. 2) Check Aetna’s online provider directory with “Riverbend Health Clinic / Dr. Okafor”
CUSTOMERRight.
AGENTSince I
CUSTOMEROkay. Got it. Just to con...
CUSTOMERYou read...
CUSTOMERThe new time and confirmation code for me? Wanna make sure I've got it right.
AGENTAbsolutely. Here are your details:
- PatientNate Brooks, date of birth 1984-03-17 - New appointment: Thursday at 4:10 PM - Location: Main Campus, Building B - Confirmation code: RVB-3M8P2
Please arrive 15 minutes early with a photo ID and your insurance card.
CUSTOMERThat's exactly right. Thanks. Alex, goodbye.
AGENTGoodbye, Mr. Brooks. You’re confirmed for Thursday at 4:10
AGENTPM, confirmation code RVB-3M8P2. Have a good day.
Home services
AGENTCopperline Home Services, this is Alex. What can I help you with?
CUSTOMERHello, Alex. I'm calling from a cafe. So sorry if it's a bit noisy. The pipe under my kitchen sink just burst. And water is spreading across the floor toward a power strip. What can you do for me?
AGENTThanks for hanging in there with me—let’s get you safe first. 1. Immediate safety steps (right now at the cafe): - If anyone is home, please call or text them right away and have them:
1.
CUSTOMERActually, I'm home right now. Could you tell me how to shut off the water?
AGENTUnderstood—you did the right thing calling. Let’s get the water and electrical risk under control first. 1) Stay safe around the power strip - Do not touch the power strip, any plugs, or outlets if the floor around them is wet. - Keep pets and kids away from that area. 2) Kill power to that area (if you can do it safely) - Go to your main electrical panel (breaker box)—often in a garage, basement, hallway, or closet. - With dry hands and standing on a dry surface, flip OFF the breaker that controls the kitchen or that outlet circuit if it’s labeled. - If you’re not sure which one it is and water is close to the strip, you can switch the main breaker off at the top of the panel to be safe. - If you’d have to walk through water to reach the panel, do not go through the water—tell me and we’ll skip this step. 3) Shut off the water main Once the electrical risk is handled or avoided, let’s stop the water:
A. Look for the main shutoff inside- Common locations: - Where the water line comes into the house (often near the front of the home). - In a basement, utility room, laundry room, or near the water heater. - The valve might look like: - A round wheel (gate valve): turn it clockwise (right) until it stops. - A lever handle (ball valve): turn the lever so it’s perpendicular (crosswise) to the pipe. B. If you can’t find that, check outside: - In many homes there’s a water meter box near the street or sidewalk (a rectangular or round lid in the ground). - If you can open it safely, there’s usually a valve you can turn a quarter turn so it’s perpendicular to the pipe to shut off. - If you’re not
CUSTOMERAlright. I'm turning the valve now. Yes. The water has stopped. What should I do next?
AGENTGood—that’s a big step. Since this started as a burst pipe and water was heading toward a power strip, we’ll treat this as an emergency plumbing visit. Before we book, a couple quick things to stabilize the situation:
1) Electrical safety - Is the floor around that power strip still wet or damp? - If yes, do not touch the strip, outlets, or any cords. - If you can reach your breaker panel while standing on a dry surface, turn off the breaker for the kitchen or that outlet circuit. If you’re unsure which, you can switch the main breaker off. - If you’d have to walk through water to reach the panel, stop and tell me instead of trying. 2) Contain the water - Put down towels, a mop, or a bucket under the sink to catch any remaining
CUSTOMERYes. The floor around the power strip is still wet. I will not touch anything there. I can reach the breaker panel safely. So I am switching off the kitchen breaker now. I will put down towels under the sink as well. What will the emergency visit cost?
AGENTPerfect—leaving that power strip alone and shutting off the kitchen breaker is exactly right. For an emergency visit today, the callout fee is:
- $149 emergency callout (this covers the technician coming out and diagnosis; parts and repairs are additional). If you’re on a Copperline Care plan, the emergency callout would be $99 instead. Do you know if you’re a Copperline Care member, or should I assume the standard $149? Once you confirm that’s okay, I’ll get a plumber booked in the earliest emergency window we have today (200 PM to 4:00 PM) and open a service
CUSTOMERI'm not a Copperline Care member. So go ahead with the standard $149. Is it possible for anyone to come before 2PM? I would prefer the earliest emergency window if that's available.
AGENTThanks for confirming. I’ve checked our emergency schedule: the earliest emergency arrival window we have today is 2:00 PM to 4:00 PM—there’s nothing available earlier than that. I’ve gone ahead and booked that for you. Here are the details:
- Visit typeEmergency plumbing visit - Callout fee: $149 (diagnosis and arrival; parts/repairs are additional, paid by card or check on completion) - Arrival window: Today, 2:00 PM to 4:00 PM - Ticket number: CPL-58210 On my side I’m seeing your info as: - Service address: 1480 Larkspur Avenue, Unit 4B Before we wrap up, I want to confirm a couple of things with you: 1) Is 1480 Larkspur Avenue, Unit 4B the correct service address? 2) Do you have any gate codes or special access instructions? 3) What’s the best callback number for the technician to reach you, and is it okay for them to text you about 30 minutes before arrival? While you’re waiting: - Keep the main water shut off. - Keep the breaker for the kitchen/outlet off and stay away from that wet area with the power strip. - Keep towels/buckets under the sink and mop up any standing water you safely can, especially near electrical outlets or cabinets. Once you confirm the address, gate info, and callback number, I’ll attach that to ticket CPL-58210 and you’ll be all set for the emergency visit this afternoon.
CUSTOMERThe correct address is 1480 Larkspur Avenue, Unit 4 D. Not 4 B. The gate code is pound 2719. My callback number is (503) 555-0139 No. 0193. Yes. It's fine for the technician to text me thirty minutes before arrival. Who will be coming, and how will I know they are on the way?
AGENTGot it, thank you for the corrections—I’ll go with what you just gave me. Here’s what I have on your ticket now:
- Service address1480 Larkspur Avenue, Unit 4D - Gate code: #2719 - Callback number: (503) 555-0139 - Ticket number: CPL-58210 - Arrival window: Today, 2:00 PM to 4:00 PM - Visit
CUSTOMERYes. That's exactly right. Thank you, Alex. Goodbye.
AGENTYou’re all set. We’ll see you this afternoon, and you’ll get a text at (503) 555-0139 about 30 minutes before arrival. If anything changes or gets worse before then—especially anything electrical near water—keep clear of the area and call us right away. Goodbye, and take care.
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
CUSTOMERYes. I am safe. I'm standing next to my car just outside a busy cafe.
AGENTI’m glad you’re safe. Is anyone injured or in need of medical help—either in your car, the other car, or nearby? If anyone might be hurt, please call 911 right away before we go any further.
CUSTOMERNo. Nobody is hurt. There is no need for medical help.
AGENTUnderstood, thanks for letting me know. Let’s get your policy pulled up so we can open a claim. Please give me:
- Your auto policy number, and - The policyholder’s last name as it appears on the policy.
CUSTOMERYes. My policy number is NG4471903. No. Sorry. Last one is two. The last name is Mehta.
AGENTThank you, Mr./Ms. Mehta. I have your policy verified under number NG4471902 for a 2021 Honda CR‑V. Next I’ll
CUSTOMERYeah. That's right. That is my card. So the other driver hit me from behind while I was stopped at a red light. The back bumper is crushed, and the rear wheel is rubbing. I cannot drive it. Police did not come. The other driver's name is Kevin Lund. Insured with Pacific Shore. And his plate is 7KX like X-ray, R214. Kindly let me know. How much do I have to pay?
AGENTI’ll walk through everything step by step and then answer how much you’ll have to pay. First, I’ve opened a claim for you and arranged a tow since the car isn’t drivable. 1) Policy and claim details - Verified policy number: NG4471902 under Mehta - Vehicle: 2021 Honda CR‑V - Claim number: NGM-904417 - Assigned adjuster: Priya Shah (she will contact you within one business day)
2) What you told me about the accident - You were stopped at a red light. - You were hit from behind by another
CUSTOMERYes. That's correct. I was stopped at the red light, and he hit me from behind.
AGENTThanks for confirming what happened. I’m glad no one was hurt. Let me answer your question about cost, then I’ll confirm your tow and next steps. 1) What you may have to pay Based on your policy coverage for policy number NG4471902:
- Collision deductible$500 That means:
- If your car is repaired under your collision coverage, you are responsible for the first $500 of covered repair costs. - If the total covered damage is more than $500, you pay $500 and insurance pays the rest (up to the value of the car), subject to the final claim decision. - Because you were rear‑ended, we will also look to recover the costs from the other driver’s insurer (Pacific Shore). If we recover from them, some or all of your deductible may be reimbursed, but that depends on their liability decision. I can’t quote
CUSTOMERBut he hit me. So I should not have to pay that. Right? Since I was stopped and he rear ended me, is the deductible still required?
AGENTYou’re right that he hit you from behind while you were stopped, and that strongly suggests he’s at fault. But yes—the collision deductible can still apply at first. Here’s how it usually works:
1) Your claim with Northgate (your own policy) - We handle your car repairs under your collision coverage. - Under that coverage, your $500 collision deductible applies, even if you were not at fault. - So you may have to pay or have $500 withheld from the repair payment initially. 2) Going after the other driver’s insurance - We will pursue Kevin Lund’s insurer, Pacific Shore, for reimbursement if they accept full liability. - If they accept fault and repay Northgate, we will then seek to get your $500 deductible reimbursed to you. - That part is not guaranteed and depends on Pacific Shore accepting liability and paying the claim. So- Yes, the deductible is still required under your policy for now. - You may get that $500 back later if the other insurance company accepts responsibility and reimburses the loss. Let me also quickly confirm what I have on file for accuracy: - Policy number: NG4471902 - Other driver’s plate: 7KXR214 - Claim number: NGM-904417 Please correct me if any of that is wrong. Do you want to: - Run the claim through your own collision coverage now (faster repairs, with the deductible), or - First try to go directly through Pacific Shore (which might avoid your deductible, but can sometimes be
AGENTslower)?
CUSTOMERYes. That is all correct. I would prefer to run the train through my own collision coverage. Get the repairs started quickly. Please arrange the tow truck. Also, can you tell me where my car will be taken and whether I will get a rental car? And will my premium go up because of this claim?
AGENTYou’re all set to run this through your own collision coverage. I’ve already started that process and
CUSTOMERHello?
AGENTI’m here with you. Let me answer everything you asked and confirm what’s already in
CUSTOMEROkay.
AGENTHere’s where things stand and what happens next. 1) Tow truck and repair
CUSTOMERRight.
AGENTI’ll keep this simple and direct. 1) Tow truck and repair shop -
CUSTOMERUh-huh. Go
AGENT2) Rental car
CUSTOMEROkay. Kindly let me know where my car will be taken.
AGENTYour car is being towed to a Northgate partner
CUSTOMERYes. I need to know if I will get a rental car, plea...
AGENTYes, you do have rental
CUSTOMEROkay. Which repair shop will it be taken to? And will I get a rent
AGENTYou’re doing everything
CUSTOMERPlease tell me about the rental car provision.
AGENTI’ll answer both clearly. 1)
CUSTOMERRight. Kindly continue.
AGENTHere are the details you asked for:
1) Repair shop
CUSTOMERSure. Go ahead.
AGENTHere are the answers, as simply as possible. 1) Where your
CUSTOMERRight. Please share the repair
AGENTHere are the specifics you asked for:
1)
CUSTOMEROkay. Where will my car be taken for repair?
AGENTThanks for your patience, and sorry for the confusion
CUSTOMERGo ahead. I'm listening.
AGENTThanks for hanging in there with me. I’ll answer your two main questions directly.
CUSTOMERNo problem. Kindly con
AGENTI’ll keep this very clear and finish the answers you’ve been asking for.
CUSTOMERSure. Please go ahead.
AGENTThanks for sticking with me. Here are the direct answers you’ve been asking for:
1)
CUSTOMERRight. Kindly let me know.
AGENTI’ll answer this directly and clearly. 1) Repair shop / tow destination Your car is being taken to our Northgate partner repair shop at 9th and Morrison. - Tow company: Eastside Towing - Destination: Northgate partner repair shop, 9th & Morrison 2) Rental car provision on your policy From your policy for NG4471902: - Rental coverage: up to $40
AGENTper day, for up to 30 days, starting today (the day the claim is opened), while your car is undrivable or in covered repairs for this claim. How that works in practice: - You are eligible for a rental car, subject to that $40/day, 30‑day limit. - If the rental costs more than $40/day, you would pay the difference. - The adjuster (Priya Shah) or the repair shop can help set up the rental with one of our preferred rental companies so it bills Northgate directly up to your limit. 3) About your premium going up Whether your premium will go up is not something I can decide here. It depends on: - The final liability decision (since you were rear‑ended, we will pursue the other driver’s insurer, Pacific Shore). - Your overall driving and claims history. - Northgate’s underwriting rules at renewal. I’m not able to quote or promise anything about future premium changes. Your adjuster, Priya Shah, can explain more once they’ve reviewed fault and the claim details. To recap the key items I must read back: - Policy number: NG4471902 - Other driver’s license plate: 7KXR214 - Your claim number: NGM-904417
Please tell me if any of that is not correct. Nextwould you like the rental arranged through the partner repair shop at 9th & Morrison, or do you prefer to contact a rental company yourself using your $40/day coverage?