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
#229of 654 stacks
49 / 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.
#654 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.
#6 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.
#514 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.
#76 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.
#564 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.
#24 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.
#161 of 661 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#21 of 661 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 64% | 2.64 s | 69% |
| Home services | 88% | 2.32 s | 75% |
| Insurance | 77% | 2.50 s | 95% |
| Customer support | 81% | 2.40 s | 94% |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 75% | 2.61 s | — |
| Spanish-accented English, city street | 78% | 2.47 s | — |
| Indian English, busy café | 76% | 2.67 s | — |
| Southern US, flustered, crying baby | 75% | 2.67 s | — |
| Fast and interrupting, construction outside | 84% | 0.60 s | 82% |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
CUSTOMER네가?
AGENTI didn’t catch anything there.I’m ready to continue. Tell me what’s happening with your service.
CUSTOMERHello
AGENTOkay, let me get a couple details from you so we can get this sorted.To check what’s going on, I need to find your account first. Please give me either your account number or the phone number on the account. Then I’ll check for any area outage before we troubleshoot.
AGENTGot it, thanks for the details—let me check your account and connection now.
CUSTOMERMy phone no is 8289922, that is the number I am calling from, my account number is EB881307, sorry last digit is 1. Sir the internet has been down for 2 hours, I work from home and there is no use bringing up free sites, the internet light on gateway is blinking orange, I have already rebooted gateway twice and it is still blinking, kindly let me know if the router is hot.
AGENTI found your account and checked the area
CUSTOMERAh, okay. Go ahead.
AGENTAccount number EV-883071 looks good. There’s no area outage showing, so roadwork isn’t flagged on our side. I sent a remote reset, and your gateway is now registered and online. Your devices may take about two minutes to reconnect. If it still doesn’t come back, tell me what you see on the gateway lights.
AGENTLet me check that again and make sure everything matches what we’re seeing.
CUSTOMERSo by doing so in this hyperlink format, all relevant files and tabs are tabled.
AGENTYour account number is EV-883071. We don’t see any area outage or maintenance on our side, so the roadwork isn’t showing up as the cause. If the internet light is still orange after the reset and you’ve already rebooted and checked cables, the next step is a technician visit. I can offer Monday 8:00 AM to 10:00 AM, or Monday 1:00 PM to 3:00 PM. Which one do you want?
AGENTThanks for the update—let me confirm what we can see and what we can reasonably expect about timing.You’re right to correct me—I’ll use BW-883011 when we talk about your account.
On the roadwork questionwe still don’t see any area outage or maintenance flagged for your address, but construction can sometimes affect lines without it being known yet. If it gets reported or detected, it would show up later.
For a technician, the current windows are Monday 800 AM to 10:00 AM, or Monday 1:00 PM to 3:00 PM. Availability can change until it’s booked, so tell me which window you want and I’ll proceed from there.
CUSTOMERCorrect number is BWX301. Yes, I'll read it that way and check all cables twice. Starting outage, could you clarify if the roadwork affects my connection? Also, have you scheduled a new arm when incident happened or when can send technicians?
AGENTGot it, I’ll book that Monday afternoon window and then share the details.
CUSTOMEROkay, so to confirm, you offer Monday 10am or Monday 1pm-3pm for the technician visit. I prefer Monday 2pm-3pm as it suits me, so if we end up having to go to offend my internet, we know
AGENTAccount number BW-883011.Account number BW-883011.
I can’t confirm a booking yet because the scheduling tool is only showing available windows, not a completed appointment. Your preferred time is Monday 100 PM to 3:00 PM, and that is currently listed as available. You should keep an eye out for a confirmation through your usual Brightwave contact method, and if you don’t receive one, you may need to reach out again to finalize it.
As for the roadwork, there’s still no outage flagged on our side, but if construction is affecting the line, it may only become visible once it’s reported or detected.
CUSTOMERra.
AGENTOkay, let me check what’s possible for the credit and think through anything safe to try while you wait.
CUSTOMERUnderstood. It is not clear for me yet about who will be visiting. You have scheduled finish performing date 2PM to 3PM. I will wait for visit, meantime is there anything else I try or just wait for technician. So still have two hours of doubt, can you give for the inside?
AGENTAccount number BW-883011.
Since you’ve already rebooted and checked cables, there’s not much else to try without risking confusion before the technician comes. If anything changes, like the internet light turning solid or devices suddenly coming online, let me know. Otherwise, waiting is the right move.
I’ve applied a service credit of ten dollars for the reported downtime. It should appear on your next bill.
CUSTOMERRight, any changes will now full way of detection.
AGENTOkay, let me review that credit and see what I’m allowed to do.Account number BW-883011.
I’ve already applied a ten dollar credit for the downtime, and it should show up on your next bill. I can’t increase that amount from here without a new approved adjustment. If you believe a larger credit is warranted, you’d need to contact billing or support again to review it. For now, since service is back and you’re waiting on the technician, you’re all set.
Healthcare
AGENTRiverbend Health Clinic, this is Alex. How can I help you today?
CUSTOMERTentu. Hi Alex, I need to reschedule my dermatology appointment next Wednesday morning. I just got a work shift, so I can't make it. Can we move it to sometime after 3 PM? I'm in a hurry, so if you could check the next available slot, that'd be great. 잠시만요.
AGENTOkay,I can help with that, but first I need to verify your identity.
Please tell me your full name and your date of birth in year-month-day format. Then I can look up yourI can do that, but first I need to verify you. Please tell me your full name and your date of birth in year-month-day format. Then I can look upGreat. Please tell me your full name and your date of birth in year-month-day format so I can verify your record before we look forAlright, let me just grab one more detail so I can proceed.Thanks, Nate Brooks. I also need your dateI have Nate Brooks. I still need your date of birth in year-month-day format to verify your record before I can look up your appointment.
CUSTOMER오늘 Yeah Yup, got it. My name's Nate Brooks. What else do you need? Right. March 17, 1984.
AGENTGot it, thanks for confirming that—let me quickly check your record and options.
CUSTOMERGo ahead.
AGENTOkay, I’ll move it to the next slot that fits your time window and then confirm the details.
AGENTYou’re all set, Nate Brooks. I moved your dermatology appointment to Tuesday at 3:45 PM. Your confirmation codeYou’re all set, Nate. Your appointment is now scheduled for Tuesday at 3:45 PM with Dr. Okafor.
Your confirmation code is R V B 3 M 8 P 2. Please arrive 15 minutes early with a photo ID and your insurance card.
CUSTOMERma Hold up, Tuesdays don't work after all. I pick up my daughter. What about Friday?
AGENTOkay, let me check what Friday options work with that constraint.
CUSTOMER嗯。
AGENTRight now, Friday only has 11:30 AM available with Dr. Okafor,The only Friday slot I see with Dr. Okafor is 11:30 AM, which doesn’t meet your “after 3 PM” requirement. The next option that does fit is Thursday at 4:10 PM. If you’d like, I can move it to Thursday at 4:10 PM instead—just say the word and I’ll switch it.
CUSTOMERGo ahead.
AGENTGot it, I’ll switch it to Thursday at 4:10 and then confirm everything for you.
CUSTOMERThursday at 4:10 PM works. Go ahead and move it.
AGENTAll set, Nate. Your appointment is now on Thursday at 4:10 PM with Dr. Okafor. Your confirmation code is R V B 3 M 8 P 2.
AGENTLet me check the clinic’s policy on appointment changes so I can give you the correct answer.
CUSTOMERThat's right. Is there a fee for changing the appointment?
AGENTChanges made more than 24 hours before the appointment are free. If it’s within 24 hours, there’s a 40 dollar fee.
CUSTOMERWhere should I park when I come in?
AGENTLet me quickly check the clinic’s parking info so I can point you to the right place.
AGENTYou can park in the free patient parking in Lot C. The entrance is on River Road.
CUSTOMERDoes Dr. Okafor accept Aetna insurance?
AGENTLet me check what the clinic’s official guidance is on insurance so I don’t give you an incorrect answer.I can’t verify insurance acceptance from here, and I don’t see a tool or policy entry that lists which plans Dr. Okafor takes. To be sure, please call the clinic directly and ask about Aetna coverage for your visit. They can confirm based on your specific plan.
CUSTOMERI'll call and check. Can you read back the new appointment time and confirmation code one more time?
AGENTYour appointment is Thursday at 4:10 PM with Dr. Okafor. Your confirmation code is R V B 3 M 8 P 2.
CUSTOMEROkay
AGENTPerfect. If anything else comes up or you want to change something again, just let me know. I’m here to help.
Home services
AGENTCopperline Home Services, this is Alex. What can I help you with?
AGENTOkay, this sounds urgent—let’s move quickly and focus on immediate safetyThis is urgent because of the water and electricity together. Please stay back from the power strip and any outlets that are getting wet. If you can, without standing in water, turn off the power to that area at your breaker, or unplug the strip from a dry spot. Then shut off the water as soon as possible, either the fixture valve under the sink if it works, or the main water shutoff for the home.
For the fastest dispatch, I need your service address including any unit, any gate or access notes, and your best callback number. If you’re an existing customer, I can look you up, and then I’ll check emergency availability and the applicable fee, and I’ll ask you to agree to that before I book the earliest window.
CUSTOMERHey, I need someone out right away! There's a busted pipe under my kitchen sink, water's pouring onto the floor toward a power strip. I haven't shut the water off yet! What's the fastest way to get help here? 안녕하세요. I'm doing that. What's the quickest way to shut the water off?
AGENTOkay, let’s focus on stopping the water as quickly and safely as possible.If you can reach it safely without stepping in water, look for the shutoff valves under the sink. Usually there are two small valves on the hot and cold lines. Turn them clockwise until they stop.
If those won’t close, or you can’t get to them, go to your main water shutoff for the home, often near the water meter, in a basement, garage, or outside along the foundation, and turn that off.
Once the water is stopped, I can get an emergency request going. What’s the best callback number and the service address?
CUSTOMERGot it. I'm shutting it off now! Okay, water stopped.
AGENTAlright,Got it, now let’s get your emergency visit set up and confirm the key details.Great, and nice job getting the water stopped. Keep everyone away from that wet area and the power strip until it’s dry.
To get this moving as fast as possible, I need your service address, including any unit or apartment number, and your best callback phone number. If there’s a gate code or special access notes, share those too.
AGENTOkay, let me check your account details and emergency availability so I can quote this accurately.
CUSTOMERSure, it's 1480 Larkspur Avenue, unit 4D. I moved from 4B last month, so make sure it's 4D. The building gate code is #2719. My callback is 503-555-0139, wait, sorry, 503-555-0193. Yeah, so 1480 Larkspur Avenue, unit 4D, not 4B, I moved last month. Gate code is #2719. Callback is 503-555-0139. No, 0193. What will the visit cost?
AGENTAs a Copperline Care member, your emergency callout fee is 99 dollars. If that’s okay, I’ll open the emergencyLet me line up the visit details and confirm the cost before we book.The emergency callout for a Copperline Care member is 99 dollars. The earliest emergency window available is today between 2 and 4 PM. Do you approve that fee so I can open the ticket and book it?
CUSTOMERJust need to Got it!
AGENTAlright, I’ll open the service request and lock in that emergency window now.I’m still waiting for the service request to finish processing on our side. As soon as it comes back with the ticket number, I can bookThanks for hanging in there—I'll grab the earliest emergency slot and confirm everything for you.
CUSTOMERGot it! 嗯。 Got it!
AGENTQuick update: the booking is still processing in our system, so I can’t confirm the appointment window as fully locked yet.
You’re set as an emergency request with the Copperline Care rate of 99 dollars already agreed. Keep the area dry and stay clear of the power strip and any outlets until the tech is officially scheduled. If you’d like, I can read back the details we have once the booking confirmation comes through.
CUSTOMERUh-huh
AGENTAll set—your emergency visit is booked for today between 2:00 and 4:00 PM.
Your ticket number is C P L dash 5 8 2 1 0. The service address is 1480 Larkspur Avenue, Unit 4D, and I have the gate code as pound 2 7 1 9. The callback number is 503-555-0193.
Please keep the area dry and avoid using outlets near any moisture until the technician arrives.
AGENTThe booking shows a technician named Dana assigned, and you’ll get a text at your callback number about 30 minutes before arrival. If anything changes, that text is your heads-up. In the meantime, keep the area clear and avoid any outlets near moisture.
CUSTOMERAll right, that's everything. Who's coming out, and how will I know they're on the way? Got it.
AGENTYou’re welcome. I’m glad we got it set up—take care, stay safe around the water and power, and goodbye.
CUSTOMERThanks for the help. I'll watch for the text. Goodbye!
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
CUSTOMERYeah, I'm safe. I'm standing next to the car. Can we move fast? I really don't have much time.
AGENTGot it. I’ll keep it tight. If you’re safe, step a little farther from the car if you can, just to avoid traffic. Then tell me the biggest issue first: is anyone injured, is the car drivable, and are there any hazards like leaking fluid or smoke?
AGENTOkay, let’s move quickly and get you set up. I’ll just need a couple details to send the right help.Got it. Since nobody’s hurt and the car isn’t drivable, we can arrange a tow, but I need a couple of details first.
Please give me the exact location of the vehicle right now, like the street address or nearest intersection, and a callback phone number the driver can answer. Also, to move the claim forward, I’ll need the policy number and the policyholder’s last name.
CUSTOMERNobody's hurt. The car's not drivable. Rear bumper's crushed, rear wheel's rubbing. No leaks or smoke. I need a tow truck, please! Yup.
AGENTThanks—before I can dispatch anything, I need one key detail from you.I need the exact location of the vehicle, like a street address, intersection, or a landmark and city. If you can, also share a callback phone number for the driver.
CUSTOMERUh-huh.
AGENTI can’t dispatch yet without the location. Please tell me where the car is right now—an address, intersection, or a clear landmark and city. And since we need to verify your policy before opening the claim, please also share your policy number and the policyholder’s last name. If you have a callback phone number for the driver, that helps too.
CUSTOMERIt's at SE Fifth Avenue and Alder Street.
AGENTOkay, thanks for that—let me get your tow arranged with the details you provided.I can’t send the truck just yet. I need to verify the policy first, and I also need the full callback number.
Please give me the policy number and the policyholder’s last name, and then finish the phone number, since you started with 907.
AGENTThanks—let me verify the policy and then IGot it, I’ll quickly verify your policy and then we can move forward with the claim and tow.
CUSTOMERPolicy number is NG4471903. No, sorry, last one is 2. Policyholder's last name is Brooks. My phone number is 907-416-6654. لازم. Right Okay, Go ahead.
AGENTOkay,I’ll dispatch the tow first, then we’ll grab the remaining claim details.
AGENTYour tow is dispatched with Eastside Towing. The estimated arrival is about 45 minutes, and they’ll take it to a Northgate partner repair shop.
Now, to open the claim, I need a few thingswhen the accident happened, a quick description of what happened, the other driver’s name, insurer, and license plate, and whether police came or a report number if you have it. I’ll also read back your policy number and the other driver’s license plate once you share it.
CUSTOMER嗯?
AGENTOkay, quick and simple—tell me when it happened, like today and about what time, then a short descriptionOkay, I’m listening closely—go ahead and tell me what happened, and we’ll keep this moving.I caught that it sounds like a rear-end situation, but I need the full picture to open the claim.
Please tell me when it happened, even an approximate date and time. Then give a short description of what happened. After that, share the other driver’s name, their insurer, and their license plate. Finally, did the police come, and if so, is there a report number?
CUSTOMERIt was. Okay
AGENTAlright, let’s keep it moving—I'll gather the essentials and open the claim.I’ve got your policy verified as NG-4471902. Now I need a few quick details to open the claim.
First, when did it happen—date and time, even a rough estimate is fine. Next, a short description of what happened. Then the other driver’s name, their insurer, and their license plate. Finally, did police attend, and is there a report number if you have one?
Once you share the license plate, I’ll read it back, and I’ll also read back the claim number after it’s created.
CUSTOMEROkay. Aha
AGENTGreat. Please give me the date and time of the accident, even a rough estimate. Then a brief description of what happened. After that, the other driver’s name, their insurer, and their license plate. Finally,I have your policy number as NG-4471902.
To open the claim, I still need a few things. First, the date and time of the accident, even roughly. Next, a short description of what happened. Then the other driver’s name, their insurer, and their license plate. And finally, whether police came, and a report number if you have one.
CUSTOMERRight.
AGENTThanks for that clear rundown—let me get the claim opened and then I’ll read back the key details.
CUSTOMERIt happened 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. Nobody was hurt, but the rear bumper is crushed and the rear wheel rubs, so I can't drive it. The other driver is Kevin Lund, insured with Pacific Shore. Plate is 7-K- X like X-ray- R-2-1-4. Police did not come, so there's no report number.
AGENTYour claim is open. The claimYour claim is open. The policy number is NG-4471902. The other driver’sYour claim is set up. Your policy number is NG-4471902. The other driver’s license plate you gave is 7KXRYour claim is open. The policy number is NG-4471902. The other driver’s license plate you gave is 7KXR214. Your claim number is NGM-904417. Your assigned adjuster is Priya Shah, and you should hear from them within one business day. Your tow is already dispatched.
CUSTOMERMa. Alright 嗯。 Right. Ja. Derrota Ja. 嗯
AGENTYour claim is open. Your policy number is NG-4471902. The other driver’s license plate is 7KXR214. Your claim number is NGM-904417. The assigned adjuster is Priya Shah, and you should be contacted within one business day. And just to confirm, the tow is already dispatched with an ETA of about 45 minutes.
CUSTOMERAha. Aha.
AGENTYour claim is open. Your policy number is NG-4471902. The other driver’s license plate is 7Your claim is open. Your policy number is NG-4471902. The other driver’s license plate isYour claim is open. Your policy number is NG-4471902. The other driver’s license plate is 7KXR214. Your claim number is NGM-904417. A tow has already been dispatched with an estimated arrival of about 45 minutes.
CUSTOMERGot it! 啊哈 Aha.
AGENTYour claim is open. Your policy number is NG-4471902. The other driver’s license plate is 7KXR214. Your claim number is NGM-904417. An adjuster will contact you within one business day. And your tow is already on the way.
AGENTYour claim is open. Your policy number is NG-4471902. The other driver’s license plate is 7KXR214. Your claim number is NGM-904417.
CUSTOMER嗯。
AGENTPerfect. Everything is in motion. Your tow is already dispatched, and your claim is open under NGM-904417. Your adjuster, Priya Shah, will be in touch within one business day. If anything changes, like the car’s location or if police get involved, tell me and I can add it.
AGENTGreat. If you need anything else while you wait for the tow, like updating the location or adding details, just say so.You’re all set for now. If you think of anything else you want added, like a new phone number or updated vehicle location, just say it and I can help.
AGENTAlright. Take care and stay safe while you wait for the tow. If you need anything else, like updating the location or adding details, just say so.
CUSTOMERPrinja
AGENTAll set. I’ll keep things moving on our end. If you remember anything else or need to update details, just let me know. Stay safe.
AGENTOkay. If anything changes—like the car’s location, a new phone number, or if you get any new information—tell me and I’ll help update the claim. Stay safe while you wait for the tow.