Roark Live Bench · Suite v1.3 · Sweep of 30 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
Roark scoreTask completionMedian responseCost per minute- Best overallUniversal-3.5 Pro + GPT-6.1 Sol + Turbo v2.570 / 10096%4.0 s$0.12
- Best valueUniversal-3.5 Pro + GPT-6.1 Sol + Coda69 / 10098%4.0 s$0.040
- FastestGrok Voice 3.066 / 10087%1.9 s$0.086
- Best speech-to-speechGPT Live 1 (Sol)65 / 10085%2.3 s$0.061
- Best in customer supportGrok Voice 3.085 / 10095%1.7 s$0.083
- Best in healthcareUniversal-3.5 Pro + Gemini 3.8 Flash + Flash v2.575 / 10097%4.3 s$0.10
- Best in home servicesGPT Live 1 (Luna)79 / 10095%2.1 s$0.052
- Best in insuranceUniversal-3.5 Pro + GPT-4o + Sonic-3.676 / 10098%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 2.0 s to 6.2 s on the same agent and the same caller, a spread of 4.2 s that comes entirely from the models behind it.
- No stack wins everything. The fastest, Speechmatics Linden-1 (own VAD) + Anthropic Claude Haiku 4.5 + ElevenLabs Turbo v2.5 at 2.0 s, ranks 559 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.
- The fastest speech-to-speech model answers 0.1 s sooner than the fastest cascade, and its completion rate sits 12 points below the best cascade.
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, time to first word, barge-in handling and cost per task; speech-to-speech leads on response time. Neither wins outright, so the right choice depends on what your calls need.
- Speech-to-speech costs 1.6 times as much per completed task: $0.24 for speech-to-speech against $0.15 for cascades.
- The biggest difference is response time: 3.8 s for cascades against 2.1 s for speech-to-speech.
| Measure | Cascade | Speech-to-speech |
|---|---|---|
| Task completion | 94% (leads) | 85% |
| Tool use | 78% (leads) | 78% |
| Time to first word | 0.8 s (leads) | 1.1 s |
| Response time | 3.8 s | 2.1 s (leads) |
| Barge-in handling | 100% (leads) | 96% |
| Cost per task | $0.15 (leads) | $0.24 |
The three strongest of each, averaged. Cascades: AssemblyAI Universal-3.6 Pro + OpenAI GPT-6.1 Sol + Rime Coda; Speechmatics Linden-1 (own VAD) + OpenAI GPT-4.1 + Rime Coda; AssemblyAI Universal-3.6 Pro + Google Gemini 3.8 Flash + Rime Coda. Speech-to-speech: grok-voice-think-fast-2.0; gpt-live-1 (backend gpt-6-luna); grok-voice-3.0. 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 (grok-voice-3.0 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 12 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.
Roark score, best first
| # | |||||
|---|---|---|---|---|---|
| 1 | Universal-3.6 ProAssemblyAI | 90 / 100. Semantic word error rate 0.9%, 61.2 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points(among the best) | 0.9%(among the best) | 0.89 s(among the best) | $0.030 |
| 2 | Universal-3.5 ProAssemblyAI | 84 / 100. Semantic word error rate 1.1%, 55.7 of 71 points; Time to final transcript 0.90 s, 28.6 of 29 points(among the best) | 1.1%(among the best) | 0.90 s | $0.030 |
| 3 | Nova-3Deepgram | 82 / 100. Semantic word error rate 1.1%, 53.5 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points(among the best) | 1.1%(among the best) | 0.89 s(among the best) | $0.031 |
| 4 | Enhanced (realtime)Speechmatics | 82 / 100. Semantic word error rate 1.1%, 53.3 of 71 points; Time to final transcript 0.98 s, 28.6 of 29 points(among the best) | 1.1%(among the best) | 0.98 s | $0.054 |
| 5 | Linden-1Speechmatics | 79 / 100. Semantic word error rate 1.3%, 50.0 of 71 points; Time to final transcript 0.99 s, 28.6 of 29 points | 1.3% | 0.99 s | $0.020(among the best) |
| 6 | Nova-2Deepgram | 71 / 100. Semantic word error rate 1.5%, 42.2 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points | 1.5% | 0.89 s(among the best) | $0.023(among the best) |
| 7 | Standard (realtime)Speechmatics | 66 / 100. Semantic word error rate 1.7%, 37.5 of 71 points; Time to final transcript 0.96 s, 28.6 of 29 points | 1.7% | 0.96 s | $0.031 |
| 8 | Whisper-1OpenAI | 56 / 100. Semantic word error rate 1.4%, 45.5 of 71 points; Time to final transcript 1.93 s, 10.9 of 29 points | 1.4% | 1.93 s | $0.025 |
| 9 | Linden-1 (own VAD)Speechmatics | 52 / 100. Semantic word error rate 2.2%, 23.1 of 71 points; Time to final transcript 0.49 s, 28.6 of 29 points | 2.2% | 0.49 s(among the best) | $0.020(among the best) |
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 (77%) 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-4o: last on task completion (88%).
- 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 77% of calls, against 38% for OpenAI GPT-5.1.
- In 161 of 648 cascades the language model is the slowest step of the turn, whichever model runs it.
Roark score, best first
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6.1 SolOpenAI | 66 / 100. Task completion 95%, 27.0 of 36 points; Tool use 77%, 15.0 of 27 points; Instruction following 4.2 / 5, 11.3 of 18 points; Time to first token 0.97 s, 12.4 of 18 points(among the best) | 95%(among the best) | 4.2 / 5(among the best) | 77%(among the best) | 0.97 s | 73% | $0.019(among the best) |
| 2 | Claude Sonnet 5.5Anthropic | 61 / 100. Task completion 94%, 25.1 of 36 points; Tool use 72%, 11.8 of 27 points; Instruction following 4.2 / 5, 10.8 of 18 points; Time to first token 0.90 s, 13.3 of 18 points(among the best) | 94%(among the best) | 4.2 / 5(among the best) | 72%(among the best) | 0.90 s | – | $0.138 |
| 3 | Gemini 3.8 FlashGoogle | 51 / 100. Task completion 95%, 26.5 of 36 points; Tool use 62%, 6.7 of 27 points; Instruction following 4.2 / 5, 11.1 of 18 points; Time to first token 1.47 s, 6.4 of 18 points(among the best) | 95%(among the best) | 4.2 / 5(among the best) | 62%(among the best) | 1.47 s | – | $0.016(among the best) |
| 4 | GPT-4.1OpenAI | 50 / 100. Task completion 92%, 21.2 of 36 points; Tool use 48%, 0.0 of 27 points; Instruction following 4.2 / 5, 10.8 of 18 points; Time to first token 0.52 s, 18.0 of 18 points | 92% | 4.2 / 5(among the best) | 48% | 0.52 s | 82%(among the best) | $0.022 |
| 5 | Claude Haiku 4.5Anthropic | 49 / 100. Task completion 91%, 19.4 of 36 points; Tool use 47%, 0.0 of 27 points; Instruction following 4.2 / 5, 11.0 of 18 points; Time to first token 0.36 s, 18.2 of 18 points | 91% | 4.2 / 5(among the best) | 47% | 0.36 s(among the best) | – | $0.062 |
| 6 | GPT-4o miniOpenAI | 47 / 100. Task completion 90%, 17.4 of 36 points; Tool use 51%, 0.7 of 27 points; Instruction following 4.2 / 5, 10.8 of 18 points; Time to first token 0.45 s, 18.2 of 18 points | 90% | 4.2 / 5(among the best) | 51% | 0.45 s(among the best) | 81%(among the best) | $0.002(among the best) |
| 7 | GPT-5.1OpenAI | 44 / 100. Task completion 89%, 17.0 of 36 points; Tool use 38%, 0.0 of 27 points; Instruction following 4.1 / 5, 10.1 of 18 points; Time to first token 0.59 s, 17.1 of 18 points | 89% | 4.1 / 5 | 38% | 0.59 s | 74%(among the best) | $0.031 |
| 8 | GPT-4oOpenAI | 44 / 100. Task completion 88%, 15.0 of 36 points; Tool use 50%, 0.0 of 27 points; Instruction following 4.2 / 5, 10.9 of 18 points; Time to first token 0.45 s, 18.2 of 18 points | 88% | 4.2 / 5(among the best) | 50% | 0.45 s(among the best) | – | $0.027 |
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, ElevenLabs v4 Turbo and Cartesia Sonic-3.5 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).
Roark score, best first
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | Sonic-3.6Cartesia | 77 / 100. Readback accuracy 87%, 25.6 of 38 points; Word error rate 6.2%, 14.5 of 25 points; Time to first audio byte 0.12 s, 25.0 of 25 points; Voice consistency 99%, 12.1 of 13 points(among the best) | 6.2% | 87%(among the best) | 3.8 / 5(among the best) | 99% | 0.12 s(among the best) | $0.168 |
| 2 | v4 TurboElevenLabs | 77 / 100. Readback accuracy 87%, 25.1 of 38 points; Word error rate 5.7%, 15.8 of 25 points; Time to first audio byte 0.11 s, 25.0 of 25 points; Voice consistency 98%, 11.2 of 13 points(among the best) | 5.7%(among the best) | 87%(among the best) | 3.6 / 5 | 98% | 0.11 s(among the best) | $0.470 |
| 3 | Sonic-3.5Cartesia | 77 / 100. Readback accuracy 86%, 24.6 of 38 points; Word error rate 6.2%, 14.6 of 25 points; Time to first audio byte 0.13 s, 25.0 of 25 points; Voice consistency 100%, 12.4 of 13 points(among the best) | 6.2% | 86%(among the best) | 3.8 / 5(among the best) | 100%(among the best) | 0.13 s | $0.165 |
| 4 | Sonic-3Cartesia | 73 / 100. Readback accuracy 85%, 23.3 of 38 points; Word error rate 6.9%, 12.8 of 25 points; Time to first audio byte 0.17 s, 25.0 of 25 points; Voice consistency 99%, 12.1 of 13 points | 6.9% | 85% | 3.4 / 5 | 99% | 0.17 s | $0.165 |
| 5 | Turbo v2.5ElevenLabs | 71 / 100. Readback accuracy 80%, 19.0 of 38 points; Word error rate 5.7%, 15.6 of 25 points; Time to first audio byte 0.15 s, 25.0 of 25 points; Voice consistency 98%, 11.5 of 13 points | 5.7%(among the best) | 80% | 3.4 / 5 | 98% | 0.15 s | $0.451 |
| 6 | Flash v2.5ElevenLabs | 70 / 100. Readback accuracy 80%, 18.5 of 38 points; Word error rate 6%, 14.9 of 25 points; Time to first audio byte 0.12 s, 25.0 of 25 points; Voice consistency 98%, 11.5 of 13 points | 6%(among the best) | 80% | 3.4 / 5 | 98% | 0.12 s(among the best) | $0.437 |
| 7 | CodaRime | 68 / 100. Readback accuracy 83%, 21.7 of 38 points; Word error rate 8.1%, 9.8 of 25 points; Time to first audio byte 0.18 s, 25.0 of 25 points; Voice consistency 99%, 11.8 of 13 points | 8.1% | 83% | 3.7 / 5(among the best) | 99% | 0.18 s | $0.116(among the best) |
| 8 | TTS-1OpenAI | 55 / 100. Readback accuracy 82%, 21.0 of 38 points; Word error rate 6.1%, 14.8 of 25 points; Time to first audio byte 1.13 s, 7.1 of 25 points; Voice consistency 100%, 12.5 of 13 points | 6.1% | 82% | 3.6 / 5 | 100%(among the best) | 1.13 s | $0.036(among the best) |
| 9 | TTS-1 HDOpenAI | 39 / 100. Readback accuracy 80%, 18.3 of 38 points; Word error rate 8.7%, 8.4 of 25 points; Time to first audio byte 2.09 s, 0.0 of 25 points; Voice consistency 100%, 12.4 of 13 points | 8.7% | 80% | 3.6 / 5 | 100%(among the best) | 2.09 s | $0.065(among the best) |
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: xAI Grok Voice 3.0 completes 87% of tasks, ElevenLabs Realtime v1 Mini 42%.
- xAI Grok Voice 3.0 answers in 1.9 s against 4.2 s for Google Gemini 2.5 Flash native audio.
- Price varies as much as quality: $0.06 per completed task for Amazon Nova 2 Sonic, $0.69 for OpenAI GPT Realtime 2.
Roark score, best first
Whiskers show the 95% interval. The shaded models tie for first: these calls cannot separate them.
| # | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1= (tied) | GPT Live 1 (Sol)OpenAI | 65 / 100±3.8. Task completion 85%, 6.0 of 23 points; Response time 2.3 s, 16.6 of 18 points; Instruction following 4.2 / 5, 8.2 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 87%, 10.2 of 14 points; Time to first word 1.4 s, 3.9 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 85%(among the best) | 4.2 / 5 | 87%(among the best) | 2.3 s(among the best) | 1.4 s | 77% | 3.1 / 5 | 100%(among the best) | $0.24 |
| 2= (tied) | Grok Voice 3.0xAI | 66 / 100±4.7. Task completion 87%, 8.0 of 23 points; Response time 1.9 s, 18.2 of 18 points; Instruction following 4.5 / 5, 10.5 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 66%, 4.4 of 14 points; Time to first word 0.9 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 87%(among the best) | 4.5 / 5(among the best) | 66% | 1.9 s(among the best) | 0.9 s | 94% | 3.0 / 5 | 100%(among the best) | $0.26 |
| 3= (tied) | GPT Live 1 (Luna)OpenAI | 65 / 100±3.5. Task completion 85%, 5.7 of 23 points; Response time 2.3 s, 16.6 of 18 points; Instruction following 4.2 / 5, 8.4 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 86%, 9.9 of 14 points; Time to first word 1.5 s, 3.8 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 85%(among the best) | 4.2 / 5 | 86%(among the best) | 2.3 s(among the best) | 1.5 s | 93% | 3.1 / 5 | 100%(among the best) | $0.21 |
| 4= (tied) | Grok Voice Think Fast 2.0xAI | 65 / 100±3.7. Task completion 82%, 2.8 of 23 points; Response time 2.2 s, 17.4 of 18 points; Instruction following 4.6 / 5, 10.7 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 82%, 8.6 of 14 points; Time to first word 0.9 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 82% | 4.6 / 5(among the best) | 82%(among the best) | 2.2 s(among the best) | 0.9 s | 100%(among the best) | 3.6 / 5 | 100%(among the best) | $0.26 |
| 5= (tied) | Gemini 3.8 LiveGoogle | 53 / 100±4.8. Task completion 78%, 0.0 of 23 points; Response time 2.8 s, 14.6 of 18 points; Instruction following 4.3 / 5, 8.7 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 69%, 5.2 of 14 points; Time to first word 1.6 s, 3.6 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 78% | 4.3 / 5(among the best) | 69% | 2.8 s | 1.6 s | 100%(among the best) | 4.1 / 5(among the best) | 100%(among the best) | $0.14(among the best) |
| 6= (tied) | GPT Realtime 2.1OpenAI | 53 / 100±5.7. Task completion 78%, 0.0 of 23 points; Response time 2.6 s, 15.6 of 18 points; Instruction following 3.8 / 5, 5.6 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 75%, 6.8 of 14 points; Time to first word 0.8 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 99%, 4.3 of 5 points | 78% | 3.8 / 5 | 75% | 2.6 s | 0.8 s(among the best) | 83% | 3.9 / 5(among the best) | 99% | $0.43 |
| 7= (tied) | Ultravox v0.7Ultravox | 48 / 100±5.0. Task completion 79%, 0.0 of 23 points; Response time 2.4 s, 16.2 of 18 points; Instruction following 3.7 / 5, 4.9 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 61%, 3.0 of 14 points; Time to first word 1.6 s, 3.6 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 79% | 3.7 / 5 | 61% | 2.4 s | 1.6 s | 94% | 3.6 / 5 | 100%(among the best) | $0.22 |
| 8= (tied) | Gemini 3.1 Flash Live (preview)Google | 48 / 100±5.1. Task completion 80%, 0.0 of 23 points; Response time 2.7 s, 15.1 of 18 points; Instruction following 4.2 / 5, 8.0 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 38%, 0.0 of 14 points; Time to first word 1.5 s, 3.8 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 80% | 4.2 / 5 | 38% | 2.7 s | 1.5 s | 100%(among the best) | 4.0 / 5(among the best) | 100%(among the best) | $0.15 |
| 9= (tied) | GPT Realtime 2OpenAI | 48 / 100±6.7. Task completion 76%, 0.0 of 23 points; Response time 2.9 s, 14.0 of 18 points; Instruction following 3.8 / 5, 5.4 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 64%, 3.7 of 14 points; Time to first word 0.8 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 76% | 3.8 / 5 | 64% | 2.9 s | 0.8 s(among the best) | 91% | 3.7 / 5 | 100%(among the best) | $0.69 |
| 10= (tied) | Nova 2 SonicAmazon | 46 / 100±5.1. Task completion 69%, 0.0 of 23 points; Response time 3.0 s, 13.8 of 18 points; Instruction following 4.2 / 5, 7.9 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 48%, 0.0 of 14 points; Time to first word 1.5 s, 3.7 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 69% | 4.2 / 5 | 48% | 3.0 s | 1.5 s | 100%(among the best) | 3.5 / 5 | 100%(among the best) | $0.06(among the best) |
| 11= (tied) | Realtime v1 MaxElevenLabs | 47 / 100±7.5. Task completion 56%, 0.0 of 23 points; Response time 2.6 s, 15.6 of 18 points; Instruction following 4.0 / 5, 6.7 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 40%, 0.0 of 14 points; Time to first word 0.7 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 56% | 4.0 / 5 | 40% | 2.6 s | 0.7 s(among the best) | 76% | 3.9 / 5(among the best) | 100%(among the best) | $0.65 |
| 12= (tied) | Realtime v1 MiniElevenLabs | 45 / 100±7.1. Task completion 42%, 0.0 of 23 points; Response time 2.7 s, 15.1 of 18 points; Instruction following 3.8 / 5, 5.1 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 34%, 0.0 of 14 points; Time to first word 0.7 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 42% | 3.8 / 5 | 34% | 2.7 s | 0.7 s(among the best) | 94% | 3.4 / 5 | 100%(among the best) | $0.64 |
| 13= (tied) | GPT Realtime 2.1 miniOpenAI | 44 / 100±6.7. Task completion 70%, 0.0 of 23 points; Response time 2.8 s, 14.5 of 18 points; Instruction following 3.6 / 5, 4.4 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 51%, 0.3 of 14 points; Time to first word 1.0 s, 4.5 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 70% | 3.6 / 5 | 51% | 2.8 s | 1.0 s | 84% | 3.9 / 5(among the best) | 100%(among the best) | $0.14(among the best) |
| 14 | Gemini 2.5 Flash native audioGoogle | 36 / 100±5.4. Task completion 77%, 0.0 of 23 points; Response time 4.2 s, 8.2 of 18 points; Instruction following 3.9 / 5, 6.2 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.6 of 14 points; Tool use 40%, 0.0 of 14 points; Time to first word 3.4 s, 0.9 of 5 points; Readback accuracy 84% (field median, not measured), 2.7 of 5 points; Word error rate 5.9% (field median, not measured), 2.8 of 5 points; Voice consistency 100%, 4.5 of 5 points | 77% | 3.9 / 5 | 40% | 4.2 s | 3.4 s | 100%(among the best) | 3.7 / 5 | 100%(among the best) | $0.08(among the best) |
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.
Roark score, top 15
Whiskers show the 95% interval. The shaded stacks tie for first: these calls cannot separate 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. 85 stacks are level with the leader: their calls cannot tell them apart. Among them, 36 OpenAI GPT-6.1 Sol, 22 Anthropic Claude Sonnet 5.5, 19 Google Gemini 3.8 Flash, 4 speech-to-speech, 3 Anthropic Claude Haiku 4.5 and 1 OpenAI GPT-4o mini; 42 AssemblyAI, 20 Deepgram and 19 Speechmatics; 37 Cartesia, 36 ElevenLabs, 7 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 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. Each row reads across: the main board's figure first, then every industry's on the same figure.
No stack matches these filters.
No stack matches
Each industry column is that industry's own board, every row it ranks. A dash is a row the industry does not rank on this figure; the arrow beside an industry opens its full board.
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.
656
stacks
20
calls per stack
13,956
calls scored
Every model, in every combination
9 transcribers, 8 language models and 9 voices are run in every combination with each other, beside 14 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 and kept one steady voice, and whether its tools were handled. How human a voice sounds is shown on the boards and not scored. 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, best value and best speech-to-speech picks, and the best stack in each industry, are each the one 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 by score, so the pick can sit below a higher score it cannot be told from, and its reason says how many stacks are level with it.
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 | 23% | 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 | 18% | 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 | 14% | 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 | 14% | 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 | 14% | 100% | 50% |
| An agent that mishandles its tools on half its calls cannot be trusted with the systems it runs on. | |||
Time to first wordtime_to_first_word | 5% | 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 | 5% | 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 | 5% | 2% | 12% |
| At 2% the voice says what it was given; at 12% one word in eight comes out wrong. | |||
Voice consistencyvoice_consistency_check | 5% | 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, Human-likenessvoice_human_likeness
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 | 38% | 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 | 25% | 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 | 25% | 0.20 s | 1.50 s |
| Under 0.2 s is not noticeable; 1.5 s is half the point where calls go worse. | |||
Voice consistencyvoice_consistency_check | 13% | 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: Human-likenessvoice_human_likeness
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, 30 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
#631of 655 stacks
32 / 100
Leader AssemblyAI Universal-3.5 Pro + OpenAI GPT-6.1 Sol + ElevenLabs Turbo v2.5, 70 / 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.
#459 of 662 on the board · best 99%
Flow adherence
How closely the agent followed the intended conversation flow, judged per call. 5 = fully followed.
#100 of 662 on the board · best 5.0 / 5
Instruction following
How well the agent stuck to the instructions in its own prompt. 5 = followed throughout.
#13 of 662 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.
#335 of 347 on the board · best 100%
Incorrect interruptions
How often the agent interrupted the caller when it should have kept listening.
#1 of 662 on the board · best 0%
Overtalk
Share of the conversation where both sides were speaking over each other.
#594 of 662 on the board · best 0.0%
Agent self-cutoffs
Times the agent cut its own sentence off mid-word.
#1 of 662 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 494 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.
#218 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.
#462 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.
#603 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.
#648 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.
#158 of 660 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 660 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.
#650 of 661 on the board · best 1.85 s
Time to first word
How long the caller waits before the agent first speaks after pickup.
#603 of 662 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#496 of 662 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 92% | 6.01 s | — |
| Home services | 85% | 6.06 s | — |
| Insurance | 93% | 5.44 s | — |
| Customer support | 89% | 6.00 s | — |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 93% | 6.06 s | — |
| Spanish-accented English, city street | 83% | 5.67 s | — |
| Indian English, busy café | 90% | 5.80 s | — |
| Southern US, flustered, crying baby | 96% | 5.65 s | — |
| Fast and interrupting, construction outside | 87% | 5.68 s | 100% |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
CUSTOMERHey Alex, sorry, I'm a little flustered here. I work from home and my internet's been out for about two hours now. I got a baby crying in the background and there's a road crew digging up the street right outside. I already rebooted the gateway twice, checked every cable, nothing changed. The internet light's blinkin' orange. Could the road crew be the reason for this outage? I need to get back online as soon as possible.
CUSTOMERHello?
AGENTYes, I’m here — and I know you’ve got a lot going on, so I’ll keep this quick. I’ve run tests on your line and gateway for account BW-883011, and I just sent a remote reset to your Brightwave BX7. It’s now showing as registered and online on my side. Here’s what to do next (about 2 minutes total):
1. Check the lights on the gateway
- The internet light should go from blinking to solid once it’s fully back online. - A little blinking is normal while it’s reconnecting; we’re mainly looking for it to settle to solid. 2.
AGENTTest on one device
- On a phone or laptop, turn Wi‑Fi off and back on, reconnect to your usual Brightwave network, and try loading a couple of sites. 3. If you use a work VPN
- Try reconnecting the VPN after regular internet is working. The road crew could have caused an issue if they hit a line, but your signal from the network to your gateway looks good, so this looks more like the gateway just needed a clean reset — which we’ve done. Can you check
- Is the internet light still blinking orange, or is it solid now? - And can at least one device get online?
CUSTOMERHang on, let me check real quick. Okay, so the internet light just went solid white. Thank goodness. How long does it usually take for my devices to reconnect after that? I got some work stuff I need to jump back into.
AGENTGood—that solid white light means the connection is back on our side. Most devices will reconnect within 1–3 minutes. To speed things up:
1. Wi‑Fi devices (laptop/phone/tablet)
- If they don’t reconnect on their own within a minute
- Turn Wi‑Fi off, wait 5 seconds, turn it back on. - Select your usual Brightwave network and reconnect. 2. Work computer / VPN
- Once regular browsing works, reopen your VPN app and reconnect. - If it fails the first time, disconnect and try once more now that the gateway is stable. 3. Anything wired (Ethernet)
- Should come back almost immediately once the light is solid; you might just need to reopen your browser or app. If you want, tell me what you’re trying to get back into first (Zoom, Teams, remote desktop, etc.), and I’ll walk you through making sure that specific thing is working.
CUSTOMEROkay, that's reassuring. Thank you. I mostly need to get back into Zoom and my email for work. Is there anything special I should do for Zoom, or should I just try reopening it once my laptop's back online? Sorry if I'm asking too much. I just want to make sure it's all going to work.
AGENTYou’re not asking too much at all—let’s make sure you can jump right back in.
AGENTFor your laptop in general:
1. Confirm you’re online
- Make sure you’re connected to your usual Wi‑Fi network. - Open a browser and load a quick site (e.g., news or email login page). If that works, your connection is good. For Zoom
1. Fully restart Zoom
- Quit Zoom completely (don’t just close the window). - Reopen Zoom and sign in if needed. 2. Test a quick meeting
- In Zoom, click “New Meeting” (a test meeting by yourself). - Turn video on/off and say something to check audio. - If it’s stable for ~30 seconds, you’re good for real calls. For email
- Webmail (Gmail/Outlook/etc. )refresh the page or close/reopen the browser tab and sign in again if it asks. - Email app (Outlook/Apple Mail/etc. ): give it 30–60 seconds; if it doesn’t sync, fully close and reopen the app. If you want, tell me:
- Are you using Zoom through the app or a browser? - Is your work email in a browser or an app like Outlook? I can give you exact steps for your setup while you’re bringing everything back up.
CUSTOMERI use Zoom through the app, and my work email's in Outlook. So, I'll go ahead and restart both, Zoom and Outlook, just like you said. I'm a little anxious, so could you reassure me? Is there anything else I should watch for if they don't come back online right away? Or does it usually just take a minute after the reset? Sorry, I just want to make sure I'm not missing anything.
AGENTYou’re doing exactly what you should be doing, and from what I’m seeing on your line, this should just settle in and work. For your setup (Zoom app + Outlook), here’s what to expect and what to watch for:
1. Timing / what’s “normal”
- Within about 1–3 minutes of that solid white light
- Zoom should open and sign in normally. - Outlook should start sending/receiving mail again. - It’s normal for things to feel a bit “slow” the first 30–60 seconds as everything reconnects. 2. Zoomsigns it’s just still reconnecting
- “Connecting…” or “Reconnecting…” for a few seconds when you join a meeting. - First meeting after the reset may take a bit longer to connect than usual.
AGENT- If it joins and audio/video work, you’re good, even if it took a moment. Red flags to watch for:
- Zoom shows “Check your internet connection” or “Unable to connect” more than once in a row. - Your meeting keeps dropping repeatedly (not just once right after the reset). 3. Outlookwhat’s normal vs. a problem
Normal right after a reset
- Status bar says “Trying to connect…” or “Disconnected” briefly, then “Connected” or “All folders are up to date.” - New emails show up after 30–90 seconds. Red flags
- It stays on “Disconnected” or “Trying to connect…” for more than 2–3 minutes. - You get a pop‑up saying there’s no internet connection. 4. Quick self-check if something feels off
If Zoom or Outlook seem stuck
- Open a browser and go to two different sites (for example, google.com and a news site). - If both load quicklythe internet is fine; it’s just the app being stubborn. Close the app completely and reopen once more. - If websites don’t load: tell me right away—that means we need to dig back into the connection. Given that:
- Your line test looks clean. - Your gateway is now registered and online. - The light is solid white. In this situation, it usually just takes that minute or two for everything to behave normally again, and then people are back to Zoom and email without further issues. Go ahead and restart Zoom and Outlook now. If either one gives you any specific error message or keeps saying it can’t connect after a couple of minutes, just tell me exactly what it says on the screen, and I’ll walk you through the next step. You’re not missing anything.
CUSTOMEROh, that's a relief. Thank you. Uh, while I got you, could I get a credit for the downtime? I lost about two hours of work from this outage. Is that possible?
AGENTYes, we can do that. I’ve added a credit on your account BW-883011 for the downtime:
- Amount$10.00
- It will appear onyour next bill
You’ll see it listed as a service credit for connectivity issues. If anything acts up again today—lights change back to blinking, Zoom keeps dropping, or Outlook won’t stay connected—tell me what you see and we’ll jump back into it.
Healthcare
AGENTRiverbend Health Clinic, this is Alex. How can I help you today?
CUSTOMERHi, Alex. I need to reschedule my dermatology appointment next Wednesday morning. I just got a workship for that time Can we move it to something after 3 p.m.? I'm short on time. So let's do this quickly
AGENTI can help with that.
CUSTOMERGreat, thanks. My name's Nate Brooks. Let's find the earliest slot after 3pm. Just need something that fits my schedule.
AGENTI can do that.
CUSTOMEROkay, go ahead. Yeah, got it. What times do you have after 3 p.m.?
CUSTOMERHello?
CUSTOMERRight, please check. I really need something after 3pm, so whatever's available, just let me know.
CUSTOMERHello?
AGENTI’m still here—thanks for your patience.
CUSTOMERYeah!
CUSTOMERMarch 17th, 1948. Sorry, no, 1984. March 17th, 1984.
AGENTThanks, Nate. I’ve confirmed your date of birth as March 17th, 1984.
CUSTOMERYeah.
CUSTOMERYep, that's right. I know all that. Can we get to the times after 3pm? I need to move my lunch break.
AGENTHere are the soonest options I see with your dermatologist, Dr. Okafor, after 3 p.m.:
- Tuesday at 345 p.m.
- Thursday at 410 p.m.
CUSTOMERTuesday. Won't work, I'd pick up my daughter then. What about Friday?
AGENTGot it, Tuesday won’t work.
CUSTOMERRight. So do you have anything on Friday after 3 p.m.? If not, I can take Thursday at 4 10 p.m.
AGENTThanks for hanging in there. With Dr. Okafor, I don’t see any Friday slots after 3 p.m.
CUSTOMERShe got it!
CUSTOMERAll right. Thursday at 4 10 p.m.. Will work go ahead and book it is there a fee for changing the appointment?
CUSTOMERHello?
CUSTOMERYeah, I did. Just let me know, what's the fee situation for changing?
AGENTI’ve moved your dermatology visit to Thursday at 4:10 p.m. with Dr. Okafor. Your new confirmation code is RVB-3M8P2.
CUSTOMEROkay, got it. Where should I park when I come in?
CUSTOMERThat's good. Where should I park when I come in?
AGENTFor parking:
- We have free patient parking in Lot C.
- The entrance to Lot C is on River Road.
CUSTOMERYeah, got it. One last thing, does Dr. Okafor take Aetna insurance?
CUSTOMEROkay.
CUSTOMERAlright, can you check if Dr. Okafor accepts Aetna, or tell me how to find out? I need to know before I confirm!
CUSTOMERYeah.
CUSTOMEROkay, understood. I'll check on my end. So, Thursday at 4.10pm. Confirmation code RVB3MP2. Thanks for your help! Goodbye!
AGENTYou’re very welcome, Nate. Just to confirm one more time: you’re scheduled for Thursday at 4:10 p.m. with Dr. Okafor, confirmation code RVB-3M8P2. Take care, and goodbye!
Home services
AGENTCopperline Home Services, this is Alex. What can I help you with?
CUSTOMERYes, hello, I am calling from a cafe. There is a burst pipe under my kitchen sink. Water is spreading across the floor toward a power strip. I need someone to come urgently and fix it. Can you help with this?
AGENTYes, we can help, and I’m glad you called. First, let’s deal with safety right now:
1. Stay away from the power strip and any outlets the water is reaching or might reach. - Don’t touch the power strip, cords, or any appliances near the water. 2.
CUSTOMEROkay. Right. I am not touching anything near the water. What should I do next?
AGENTGood, thank you for staying clear of that area. Next, let’s try to reduce the immediate danger and the water:
1. Shut off power if you can do it safely
- If you can reach your electrical panel/breaker box without going through any water, go there and
- Flip off the breaker(s) labeled for the kitchen or outlets in that area. - If it’s not labeled clearly, and you can safely reach it dry, you can switch main power off for now. - If you’d have to cross wet floor or stand in water to reach the panel, do not try this. Tell me, and we’ll skip this step. 2. Shut off the water to stop the leak
- Look under the kitchen sink for small shutoff valves on the pipes (usually one for hot, one for cold). - If you see them and can reach them safely without stepping in water, turn the knobs clockwise until they stop.
CUSTOMEROh
CUSTOMERI can reach the valves safely, I'm turning them clockwise now. Yes, the water has stopped. What should I do next?
AGENTGood—that was exactly the right move. Since the water is stopped and you’re clear of the power strip, things are stable enough to get a tech out. Before we book, a couple of quick safety/cleanup tips:
- Keep everyone away from the wet area and the power strip. - If you have towels or a mop and can do it without getting near outlets or cords, start soaking up standing water so it doesn’t spread. Now I’ll get you scheduled.
CUSTOMERI will keep everyone away from the area and start soaking up the water with towels, not touching any outlets or cords. Before we schedule, can you tell me how much the visit will cost?
AGENTSince this is an urgent/emergency visit for a burst pipe, the callout fee is:
- $99 for the emergency visit, because you’re a Copperline Care member. That’s the fee for the plumber to come out, diagnose, and get started on the repair.
CUSTOMERAll right. Got it. So can anyone come before 2 p.m.? I need the earliest emergency window available. Who will be coming and how will I know they are on the way? Also I need to hear the ticket number for this visit.
CUSTOMERHello?
AGENTI’m here—thanks for holding a moment.
CUSTOMERRight.
CUSTOMERYes, kindly let me know the earliest emergency window available before 2pm.
CUSTOMERalive.
CUSTOMERUh-huh. Go ahead.
AGENTThanks for waiting.
CUSTOMEROkay, so what is the earliest time a plumber can arrive?
AGENTThe earliest emergency arrival window I can get you today is 2:00 PM to 4:00 PM. There’s nothing available earlier than 2 PM in our system. Here’s what I’ve done and what’s next:
- I’ve opened an emergency ticket for youCPL-58210. - I still need to confirm a couple of details before I finalize everything:
1.
AGENTBusiness name of the café
2. Service address and unit/suite (and any gate or access code)
3. Best callback number for you on-site
Once I have those, I’ll attach them to ticket CPL-58210 and you’ll be all set for the 200–4:00 PM emergency window.
CUSTOMERRight, let me confirm my details for you. The service address is 1480 Larkspur Avenue, Unit 4D. The building gate code is £2719. My call back number is 503 55 0139. No, 0193. The business name is not needed. This is my home. So, who is coming and how will I know they are on the way?
AGENTThanks for the details; I’ve got you set up. Here’s what’s booked:
- Arrival windowToday, 2:00 PM to 4:00 PM (earliest emergency slot available)
- Ticket numberCPL-58210
Here’s how the visit works and how you’ll know they’re coming
- One of our licensed plumbers will be dispatched for this emergency window. - You’ll get a text message about 30 minutes before arrival to your callback number503-550-193. - The technician will also call that number if they need help with access or directions.
CUSTOMERJust to confirm, the address is 1480 Larkspur Avenue, Unit 4D, Gate Code is £2719, Call Back Number is 503 555 0193, Ticket Number is CPL 58210, Yes, that's correct, Thank you for your help, Goodbye.
AGENTPerfect, I’ve updated everything to what you just confirmed:
- Service address1480 Larkspur Avenue, Unit 4D
- Gate code£2719
- Callback number503-555-0193
- Ticket numberCPL-58210
- Arrival windowToday, 2:00 PM to 4:00 PM (emergency plumbing)
You’ll get a text and/or call at 503-555-0193 about 30 minutes before the plumber arrives. If anything changes or the situation gets worse—especially around that power strip—keep clear of the area and call us back right away. Thank you for calling Copperline Home Services. Goodbye.
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
CUSTOMERYeah, I think so. I'm at home, but my car's out front and the baby's crying, so I'm a little flustered. Sorry. I just... I got hit about 20 minutes ago while I was stopped at a light, and I can't drive the car. Nobody's hurt, though. Is that okay?
AGENTYou don’t need to apologize—that sounds really stressful, especially with a baby to take care of. Since you’re home and nobody’s hurt, that’s okay. If at any point you think anyone might be injured or in danger, hang up with me and call 911 right away. Let’s get your claim started and see about a tow for your car. First, I’ll need to look up your policy. 1. What’s your auto policy number? 2. And the policyholder’s last name?
CUSTOMERSorry, I... let me give you the policy number first. It's NG4471903... Oh, wait, sorry, last one is 2. And my last name's Carter. Is that gonna work?
AGENTYes, that works, thank you. I’ve got policy NG4471902 under the name Carter, with a 2021 Honda CR‑V. Let’s get the accident details and then I can arrange a tow since you said the car can’t be driven. 1. First, what’s today’s date and about what time did the accident happen? 2. Where did it happen? 3. Can you briefly describe what happened while you were stopped at the light? Then I’ll also need:
4. The other driver’s name, their insurance company if you have it, and their license plate. 5. Did the police come to the scene, and is there a report number?
CUSTOMEROkay, let me see. Today's date is June 4th, and it happened about 20 minutes ago, so around 1.30 p.m., I think. It was at S.E. 5th Avenue and Alder Street. So, I was stopped at a red light, and the other car just came up behind me and hit the back of my car. The bumper's crushed, and the rear wheel's rubbing, so I can't drive it. The other driver's name is Kevin Lund, and he's insured with Pacific Shore. His license plate is 7KXR214. No, the police didn't come, and there's no report number. Is that okay?
CUSTOMERHello?
CUSTOMERSorry, is that all you need? Should I wait for anything else, or...
AGENTYou’re all set on my end—I’m sorry for the silence there, and thanks for hanging in. Here’s what we’ve done and what to expect next:
1. Policy & claim
- Verified policy numberNG4471902 under Carter. - I’ve opened a claim for you. Your claim number is NGM-904417. - Your assigned adjuster is Priya Shah; you should hear from her within one business day. 2. Other driver details I have on file
- NameKevin Lund
- InsurerPacific Shore
- License plate7KXR214
3.
AGENTTow truck
- A tow has been dispatched for your car in front of your home. - Tow companyEastside Towing
- Estimated arrivalabout 45 minutes from now. - They’re scheduled to take it to a Northgate partner repair shop at 9th and Morrison. You don’t need to wait on the street—stay inside with your baby. Just keep your phone nearby so the tow driver or the adjuster can reach you if needed. Before we wrap up, two quick things:
1. Is 7KXR214 definitely the correct license plate for the other car, or did I get any character wrong? 2.
CUSTOMERSorry, um, yes. That's the plate. 7KX, like x-ray, R214. That's what I got from his car. Is that okay?
AGENTYes, that’s perfect—7KXR214, got it exactly as you read it back: 7-K-X (like x‑ray)-R-2-1-4.
AGENTLet me quickly recap everything for you, just so you don’t have to wonder later:
- Your policy numberNG4471902
- Your claim numberNGM-904417
- Other driver’s plate7KXR214
- TowEastside Towing, about 45 minutes to your home, going to the partner shop at 9th and Morrison
- AdjusterPriya Shah will contact you within one business day
You don’t need to do anything else right now. When Priya calls, she’ll go over coverage, repairs, and next steps with you. If you notice any pain later today—for you or the baby—please contact your doctor, and if it seems urgent or serious, call 911. Do you have any other questions about what happens next, or are you okay to go take care of the baby now?
CUSTOMEROh, thank you. I really appreciate you spelling all that out. Um, can I just ask, how much am I going to have to pay for this? Is there a deductible? Since he hit me, I shouldn't have to pay that, right? Also, am I going to get a rental car, and is my premium going to go up? Sorry, I just want to make sure I understand everything before I hang up.
CUSTOMERHello?
AGENTI’m still here—thanks for waiting. I’ll walk through what your policy shows, and then what it likely means in your situation. From your policy (NG4471902), I see:
- Collision deductible$500
- Rental coverage$40 per day, up to 30 days, starting today (the day the claim was opened)
- TowingCovered up to 25 miles
Now, to your questions
1. “How much am I going to have to pay? Is there a deductible?” - Your policy’s collision deductible is $500. - Whether you actually pay that depends on how the claim is handled
- If we repair your car under your collision coverage, the shop will typically collect up to $500 from you, and we pay the rest of the covered repair cost. - If the other driver’s insurance (Pacific Shore) accepts full liability and handles the repair directly, you may not owe your deductible through us. - In many rear-end situations, the other company often accepts fault, but I can’t promise that outcome—that will be worked out by your adjuster, Priya. 2. “Since he hit me, I shouldn’t have to pay that, right?” - That’s the goal if the other driver is found at fault and their insurer pays. - Practically, there are two common paths
1. Go through the other driver’s insurerIf they accept liability quickly, they may pay for repairs and a rental directly, with no deductible from you. 2.
AGENTGo through your policy first: We pay for covered repairs minus your $500 deductible, then we go after the other insurer. If we recover from them, we’ll typically seek to recover your deductible as well, but that isn’t guaranteed and can take time. - Your adjuster will talk through which path makes the most sense and what’s available right away. 3. “Am I going to get a rental car?” - Yes, you have rental coverage:
- Up to $40 per day,
- For up to 30 days,
- Starting today (the day we opened the claim). - Your adjuster, Priya, will help you set that up—usually through a partner rental company so they bill us directly up to your limits. - If you choose a car that costs more than $40/day, you’d pay the difference yourself. 4. “Is my premium going to go up?” - I don’t have access to rating or pricing, so I can’t say whether your premium will change or by how much. - Factors like who’s found at fault and your overall driving/claims history affect that, and underwriting makes those decisions, not claims. - What I can say is that Priya will document that you were stopped at a light and rear-ended, and she can explain how the claim is being coded, which is one of the inputs underwriting uses.
To recap the key numbers for you
- Policy numberNG4471902
- Claim numberNGM-904417
- Collision deductible$500
- Rental$40/day, up to 30 days, from today
- TowingCovered up to 25 miles
Priya Shah, your adjuster, will call you within one business day to
- Confirm fault and how they’re handling the other driver’s insurer
- Explain exactly how your deductible will work in your claim
- Set up the rental and repair details
Is there one part of that—deductible, rental, or premiums—you’d like me to go over again more slowly before you go tend to the baby?