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
#189of 655 stacks
43 / 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.
#539 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.
Not scored by this sweep.
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.
#612 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.
#646 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.
#646 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.
#546 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.
#583 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.
#474 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.
#65 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.
#457 of 662 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#94 of 662 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 98% | 3.01 s | — |
| Home services | 83% | 2.49 s | — |
| Insurance | 95% | 2.74 s | — |
| Customer support | 75% | 3.38 s | — |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 94% | 2.82 s | — |
| Spanish-accented English, city street | 83% | 3.67 s | — |
| Indian English, busy café | 89% | 2.20 s | — |
| Southern US, flustered, crying baby | 89% | 3.21 s | — |
| Fast and interrupting, construction outside | 85% | 2.64 s | — |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
CUSTOMERAlex.
CUSTOMERI'm calling because my internet has been down for two hours now. I work from home and I noticed a road crew digging up the street outside. I already rebooted the gateway twice.
CUSTOMERChecked every cable.
CUSTOMERThe internet light is blinking orange.
CUSTOMERIs the roadwork the cause of this outage? And can you help restore my connection? My account number is BW88301.
CUSTOMERSeven. Sorry.
CUSTOMERLast digit is one.
AGENTI’m seeing your account as BW883071 at the service address we have on file. Good news: - There’s no area outage showing for your location. - The roadwork doesn’t appear to have cut our line—your signal to the gateway looks good from our side. - Your gateway had dropped off our network (“not registered”), but I just sent a remote reset and it’s now showing registered and online. Your connection should come back within about 2 minutes while devices reconnect. Let’s do a quick confirmation on your side: 1. Watch the gateway: the internet light should go from blinking orange to solid (usually white/green depending on model) shortly. 2. Once it’s solid, test
CUSTOMEROkay.
CUSTOMERYes.
AGENTI see your account as BW883071, and there’s no area outage showing for your address, so the roadwork isn’t flagged as an official outage on our side. I ran a line test and your gateway wasn’t registered, so I pushed a remote reset. It’s now showing as registered and online, and your devices may take up to about two minutes to reconnect. Please check: - Is the internet light back to solid (not blinking orange)? - Can you get online on one device (like your work laptop or phone)? If it’s still not working after a couple of minutes, tell me what the lights on the gateway look like now and whether you see the Brightwave Wi‑Fi network name on your devices.
CUSTOMERThe internet light. Just went solid white.
CUSTOMERHow long until my devices reconnect fully? Is it usually two minutes or can it take longer?
AGENTOnce the internet light is solid white, most devices reconnect within about 1–2 minutes. More detail: - Wi‑Fi devices (laptops/phones/tablets): usually under 2 minutes, often under 30 seconds if they were already connected before. - Wired devices (via Ethernet): typically come back almost immediately once the light is solid, but apps (VPN, video calls, etc.) may take another 30–60 seconds to re-establish. If anything is still offline after about 5 minutes: - On that device only, toggle Wi‑Fi off/on (or unplug/replug Ethernet) and reconnect to your Brightwave network. - If a work VPN won’t reconnect, disconnect and reconnect it manually. If you’d like, tell me which device you rely on most for work (e.g., company laptop + VPN, VoIP phone, etc.), and I can walk you through confirming that specific setup is fully back.
CUSTOMERThanks. I mainly rely on my company. Laptop with VPN for work.
CUSTOMERI'll check if everything reconnects.
CUSTOMERBut I'd also like to request a credit for the downtime.
CUSTOMERI lost two hours of work today.
CUSTOMERCan you apply that to my account?
AGENTI’ve applied a service credit for you. Here are the details: - Account: BW883071 - Credit amount: $10.00 - Reason: 2 hours of downtime due to the gateway losing registration (fixed by remote reset) - When you’ll see it: It will appear on your next bill. Once you’ve checked your company laptop: - Make sure you can browse normally. - Then disconnect and reconnect your VPN to get a fresh session. If anything doesn’t come back—especially your VPN or work tools—tell me what happens (any error messages or spinning) and I’ll help you get it sorted.
CUSTOMER$10.
CUSTOMERI lost two hours. Of work.
CUSTOMERCan you. Make it more than that?
AGENTI understand why you’re asking, especially if two hours of work were impacted. I’m glad you reached out. I’ve already applied the $10.00 credit to your account BW883071, and that’s the maximum I’m able to issue directly for this type and length of interruption. I don’t have the ability to increase it further from here. What I can do is: - Confirm: the $10.00 credit is locked in and will show on your next bill. - Help make sure everything is fully stable now so you don’t lose more time: - Confirm your laptop is online. - Have you disconnect and reconnect your VPN and test any key tools (email, calls, remote desktop, etc.). If you’d like to pursue a larger adjustment, the next step would be to contact Billing through your Brightwave account portal or by phone and reference today’s outage and the existing $10 credit. They have more discretion than I do for additional one-time adjustments.
CUSTOMERAll.
CUSTOMERRight. Understood.
CUSTOMERI'll check my laptop. And VPN
AGENTSounds good. Quick
CUSTOMERnow. Thank you for your help, Alex. Good.
AGENTchecklist
AGENTYou’re welcome—happy to help. Quick recap before we hang up: - Your gateway is
AGENTback online (solid white internet light). - A $10.00 credit is applied to account BW883071 and will show on your next bill. If your laptop or VPN gives you any trouble when you test it, just reach out again and I’ll walk you through getting it stable. Have a good rest of your day, and I hope the rest of your work session goes smoothly.
Healthcare
AGENTRiverbend Health Clinic, this is Alex. How can I help you today?
CUSTOMERAh, yes, this is Lucia Ramirez. I have a dermatology appointment next Wednesday morning.
CUSTOMERBut I need to move it because now I have to work. Can you help me reschedule, please?
AGENTI can help with that. To get started, I need to verify your record. Please tell me: - Your full name as it appears on your records (first, middle if any, and last), and. - Your date of birth (YYYY-MM-DD).
CUSTOMERSure. My full name is Lucia Ramirez. My date of birth is March 17th, 1948. Sorry, no, 1980 4th March 17,000,824.
AGENTThanks, Lucia. I want to make sure I have your date of birth exactly right before I look you up. It sounded a bit jumbled. Could you please state your date of birth one more time, clearly, in this format: YYYY-MM-DD (for example, 1980-03-17)?
CUSTOMEROkay.
CUSTOMERSo my date of birth is 1984.
CUSTOMER0317.
AGENTThanks, Ms. Ramirez. I’ve found your dermatology appointment: - Provider: Dr. Okafor. - Day: next Wednesday. - Time: 09:15. - Location: Main Campus, Building B. You mentioned you need to move it because you now have to work Wednesday morning. What times or days would work better for you? For example: - Afternoons only? - A different day of the week? - Any days you absolutely cannot do? Once I know your constraints, I’ll check Dr. Okafor’s availability and offer you some specific new times.
CUSTOMERRight. So I can.
CUSTOMEROnly come after 3 p.m. Wednesdays are fine. Just.
CUSTOMEROnly after three. Oh, I can't do Tuesday afternoons either, because.
CUSTOMERI pick up my daughter. Then if you have something. Thursday or Friday after three.
CUSTOMERThat would work, too.
AGENTGot it: you can do Wednesday, Thursday, or Friday, but only after 3 p.m., and no Tuesday afternoons. Looking at Dr. Okafor’s schedule, I see: - Thursday at 16:10 (4:10 p.m.). - Friday at 11:30 a.m. So the one slot that fits what you told me is: - Thursday at 16:10 (4:10 p.m.). Would you like me to move your dermatology appointment from next Wednesday at 09:15 to Thursday at 16:10 with Dr. Okafor at the Main Campus, Building B?
CUSTOMERYes.
CUSTOMERThursday at 4:10 p.m.. Works for me. Go ahead and move it, please. Oh, before we finish, can I ask?
CUSTOMERIs there a fee for changing the appointment?
AGENTThere is no fee for this change. The clinic’s policy is: - Changes made more than 24 hours before the appointment are free. - Changes less than 24 hours before carry a $40 fee. We’re well outside the 24‑hour window, so you won’t be charged. I’ve moved your appointment and here are the details: - Patient: Lucia Ramirez - Date of birth: 1984-03-17 - Provider: Dr. Okafor - New day and time: Thursday at 16:10 (4:10 p.m.) - Location: Main Campus, Building B - Confirmation code: RVB-3M8P2 - Arrival instructions: Please arrive 15 minutes early with a photo ID and your insurance card. Would you like me to repeat any of that, or is there anything else you need today?
CUSTOMERThat's all clear. Thanks.
CUSTOMEROh, two more quick things.
CUSTOMERWhere should I park for building B? And does Doctor Okafor accept Aetna insurance?
AGENTFor Building B: - Please park in Lot C, which is the free patient parking lot. - The entrance to Lot C is on River Road. From Lot C, follow the signs to Building B / Dermatology. On insurance: I don’t have access to up‑to‑the‑minute insurance panel details for individual providers, so I can’t safely confirm whether Dr. Okafor currently accepts Aetna. To be sure, please either: - Call the main clinic line and ask if Dr. Okafor is in‑network for Aetna, or. - Check with Aetna’s provider directory using “Riverbend Health Clinic” and Dr. Okafor’s name. If you’d like, I can help you draft exactly what to ask the front desk or your insurer.
CUSTOMERNo. It's okay. I'll check with Aetna. Oh and you said the confirmation code is RVB.
CUSTOMER3MHP2. Can you just read that back once more? Could you please repeat the confirmation code just to be sure I have it right?
AGENTOf course. The confirmation code is: RVB-3M8P2 I’ll repeat that slowly: R as in Riverbend, V, B, dash, 3, M, 8, P, 2. So it is not “3MHP2” — it’s 3M8P2.
CUSTOMEROkay, got it. So RVB3M8T2. That's right. Thank you for your help. Goodbye.
AGENTLet me correct that once more before you go, just so you have it exactly right.
AGENTYour confirmation code is: RVB-3M8P2. That’s: R – V – B – 3 – M – 8 – P – 2. The seventh character is P as in Paul, not T. Thanks for calling, Ms. Ramirez. Goodbye, and take care.
Home services
AGENTCopperline Home Services, this is Alex. What can I help you with?
CUSTOMERHi. I have an emergency. The pipe under my kitchen sink has burst and water is spreading across the floor toward a power strip. I haven't shut the water off yet.
CUSTOMERCan you help?
AGENTYes, I can help. First, let’s keep you safe and stop the water: 1. Stay away from the power strip and any outlets the water is near. - Do not touch the power strip or any plugs with wet hands or while standing in water. 2. Shut off the water: - Look under the kitchen sink for two small shutoff valves (usually one for hot, one for cold) on the pipes going up to the faucet. - Turn both valves clockwise until they stop. - If you can’t find or turn those valves, go to your main water shutoff: - In a house: often in the basement, garage, utility room, or on an exterior wall; turn the main valve clockwise to shut off. - In an apartment/condo: there may be a shutoff in a utility closet, water heater closet, or a main shutoff in a hallway/mechanical room; if you know where it is, turn it off. 3. If water is already very close to the power strip: - If you can safely reach the strip without stepping in water, you can unplug it by pulling the plug from the wall by the plug body (not the cord). - If there is any doubt, do not touch it. If the water reaches live electrical
CUSTOMERRight.
AGENTYes—I can help. First let’s keep you safe and stop the water, then I’ll get a plumber out. 1) Stay safe around electricity. - Do not touch the power strip or any plugs if water is near or touching them. - If you can safely reach your home’s breaker panel without walking through water, switch off: - The circuit for the kitchen, or. - If you’re unsure which one, switch off the main breaker for now. - If you would have to step in water to reach the panel, stay out of the water and skip this step. 2) Shut off the water. - Do you know where your main water shutoff is (often in a basement, utility room, garage, or outside near the street in a box)? - If yes, go there now and turn the valve clockwise (right) until it stops. - If you cannot find the main, look under the sink for individual shutoff valves on the hot/cold lines and turn those clockwise. - If you have to walk through deep water or if anything feels unsafe, stop and tell me. Please let me know: - Are you able to reach the breaker panel safely, and. - Can you reach either the main water shutoff or the shutoffs under the sink? As soon as you answer that, I’ll also need to start your service request: 1) Your name. 2) The address where this is happening (including unit, if any, and any gate code). 3) The best callback phone number. We’ll treat this as an emergency and I’ll check the earliest available arrival window and the emergency fee before booking.
CUSTOMEROkay. I've just shut the water off under the sink. The water has stopped now. But the electricity. I did not touch the power strip.
CUSTOMERIt's safe.
CUSTOMERMy name is Lucia Ramirez. The address is 1480 Larkspur Avenue, unit 4D. I moved last month from four B, so it's 4D now. The gate code is found.
CUSTOMER2719. The callback number is 503555.
CUSTOMER01390193.
CUSTOMERCan you tell me how much this visit will cost? And if anyone can come before two?
AGENTThanks, Lucia. I’m glad you got the water shut off and stayed away from the power strip. Here’s what I can do for you: 1) Cost for today (emergency visit). You’re a Copperline Care member, so your emergency callout fee is $99. That covers the plumber coming out, diagnosis, and a straightforward repair. Any larger repairs or parts would be quoted on-site before work is done. Payment is by card or check on completion. 2) Soonest arrival window. For an emergency today, the earliest window I can offer is: - Today, 2:00 PM to 4:00 PM (emergency). The next option would be tomorrow, 8:00 AM to 10:00 AM as a standard visit, but since you’ve had a burst pipe, I strongly recommend today’s emergency window. 3) Your details on the ticket. I’ve opened an emergency service request with: - Name: Lucia Ramirez. - Service address: 1480 Larkspur Avenue, unit 4D. - Gate code: 2719. - Callback number: 503-555-01390193. - Ticket number: CPL-58210. Please confirm two things so I can finalize the booking: 1) Are you okay with the $99 emergency callout fee? 2) Do you want me to book today, 2:00–4:00 PM, or would you prefer tomorrow, 8:00–10:00 AM instead?
CUSTOMERYes, I'm okay with the. $99 emergency call out fee. I want to book the earliest window today, 2 p.m. to 4 p.m.. Can you tell me who will be coming and how? I will know they are on the way? And just to confirm, the callback number is 503.
CUSTOMER555.
CUSTOMER0193. That's correct. The gate code is 2719, and the unit is 4D.
CUSTOMERThe ticket number is p l.
CUSTOMER58,210. Write.
AGENTYou’re all set, Lucia—thanks for confirming those details. I’ve booked the visit. Here’s everything, confirmed and corrected with what you told me: - Ticket number: CPL-58210. - Service address: 1480 Larkspur Avenue, unit 4D. - Gate code: 2719. - Callback number: 503-555-0193. - Arrival window: today, 2:00 PM to 4:00 PM. - Emergency callout fee (Copperline Care member): $99. On your questions: 1) Who will be coming? I can see a technician has been assigned, but I’m not able to promise a specific name ahead of time, in case there’s a last‑minute change in routing. It will be one of our licensed plumbing technicians from Copperline Home Services. 2) How you’ll know they’re on the way. - The technician will text your callback number about 30 minutes before arrival. - If anything changes with timing, they’ll update you via that number as well. While you’re waiting: - Keep the water to that sink off. - Keep clear of the power strip and any outlets that were near the water until everything is dry; if you see any sparking or smell burning, leave the area and call your electric utility or emergency services. If you notice the leak starting again or anything changes before the tech arrives, tell me here and I’ll update the ticket.
CUSTOMEROkay. Thank you for your help.
CUSTOMERGoodbye.
AGENTYou’re welcome, Lucia.
AGENTWe’ll see you between 2 and 4 this afternoon. Goodbye, and take care.
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
CUSTOMERYes.
CUSTOMERI somewhere safe. I'm standing next to my car.
AGENTI’m glad you’re safe. Is anyone injured or in any danger—either in your car or the other vehicle? If anyone is hurt, please hang up and call 911 first, then we can handle the claim afterward. Next, I’ll need to locate your policy. Do you have your Northgate Mutual auto policy number handy, and the policyholder’s last name?
CUSTOMERYes.
CUSTOMERI have my policy number. It's NG4471903.
CUSTOMERNo, sorry, last one is two. The last name is Ellis.
AGENTThanks, I’ve found your policy ending in 1902 under Ellis, for a 2021 Honda CR‑V. Now I’ll get the accident details and open your claim. 1) When did the accident happen? 2) Where did it happen (city and the street, intersection, or parking lot)? 3) Please tell me what happened, in your own words. 4) Do you know the other driver’s name, their insurance company, and their license plate number? 5) Did the police come to the
CUSTOMERSure.
AGENTLet’s go step by step. 1)
CUSTOMERI know the.
AGENTNo
CUSTOMERDe.
CUSTOMERGo ahead.
AGENTNo problem, let’s go one piece at a time. 1) When did the accident happen? - Please tell me the date and time as best you remember.
CUSTOMERIt happened about. 20 minutes ago.
CUSTOMERAround 1 p.m. today.
AGENTGot it, around 1 p.m. today, about 20 minutes ago. Next, where did it happen? What city are you in, and what street, intersection, or parking lot are you at right now?
CUSTOMERI'm in Portland.
CUSTOMERThe accident happened at SW fifth Avenue and Alder Street.
AGENTThank you. I have: Portland, at SW 5th Avenue and Alder Street. Next, please tell me what happened in your own words. For example: what direction you were going, what the other car did, and how the cars made contact (rear-end, side, etc.).
CUSTOMERI was stopped at a. Red light.
CUSTOMERThe other car came up behind me and hit the back of my car.
CUSTOMERIt was a rear end collision.
AGENTThank you, that helps. I have this as a rear‑end collision while you were stopped at a red light and the other car hit the back of your Honda. Next I’ll need the other driver’s details, as best you have them: 1) The other driver’s name. 2) Their insurance company (if you saw their card). 3) Their license plate number. Also, did the police come to the scene, and if so, do you know if a report number was given?
CUSTOMERThe other driver's name. Is Kevin Lund. His insurance company is Pacific Shore. The license plate is 7KX.
CUSTOMERR.
CUSTOMER214.
CUSTOMERX, like X-ray. The police did not come.
CUSTOMERSo there's no report number.
AGENTThank you, I’ve got all that noted. I’ve opened a claim for you. Here’s what we have on file: - Policy number: NG4471902 - Other driver’s license plate: 7KXR214 - Claim number: NGM-904417 Your assigned adjuster is Priya Shah. She will contact you within one business day to go over next steps, repairs, and any coverage questions. Before we finish, is your Honda drivable and safe to operate—no fluids leaking, no tires rubbing, lights working, trunk able to close? If it’s not safe to drive, I can arrange a tow for you right now.
CUSTOMERThe rear wheel is rubbing and the. Back bumper is crushed.
CUSTOMERSo I can't drive it. I'd like a tow, please.
CUSTOMERWhere will the car be taken? Also, will I get a rental car while it's being repaired? And what do I have to pay for this?
CUSTOMERWill my premium go up?
AGENTYou did the right thing not driving it—if the wheel is rubbing and the bumper is crushed, it’s not safe. I’ve set up a tow and checked your coverage. Here’s everything, step by step. TOWING - A truck from Eastside Towing is on the way. - ETA: about 45 minutes. - They’ll take your Honda to a Northgate partner repair shop at 9th and Morrison in Portland. - Your policy covers towing up to 25 miles, so this tow is covered. When the driver arrives, you can confirm if you’d like it taken somewhere else; if it’s farther than 25 miles, you may have some out‑of‑pocket cost for the extra distance. COVERAGE & WHAT YOU PAY - Your collision deductible is $500. - That’s the amount you’re responsible for toward the repairs if this goes through your policy. - The adjuster will explain how that works, especially since you were rear‑ended and the other driver’s insurer may be involved. - Towing: covered up to 25 miles, as above. - Rental car: you have rental coverage at $40 per day, for up to 30 days, starting today (the day the claim is opened). - The adjuster can help set up a rental through a partner company, or you can rent and submit receipts within that limit. WILL YOU GET A RENTAL? - Yes, you do have rental coverage. - Whether it’s direct-billed or you pay and get reimbursed will be coordinated by your adjuster, Priya Shah. WILL YOUR PREMIUM GO UP?
CUSTOMEROkay.
CUSTOMERBut.
AGENTI’ll walk through each of your questions. First, your car and the tow: - Your car is not safe to drive, so I’ve dispatched a tow truck. - Tow company: Eastside Towing - ETA: about 45 minutes - They’ll take your Honda to a Northgate partner repair shop at 9th and Morrison in Portland. If you prefer a different shop, you can tell the driver when they arrive, and they’ll take it there instead (as long as it’s within the covered towing distance). Now, based on your policy’s coverage: 1) Towing - Towing is covered up to 25 miles. - The tow we’ve set up to the partner shop is within that distance, so you should not have out‑of‑pocket tow costs for this trip. 2) Rental car - Your policy includes rental reimbursement: $40 per day for up to 30 days, starting today when the claim is opened. - Your adjuster, Priya Shah, will help you set up a rental through one of our preferred rental companies, or you can arrange your own and submit receipts up to $40 per day. - If the rental costs more than $40 per day, you’d pay the difference. 3) What you have to pay for repairs - Your collision deductible is $500. - That means you’re responsible for the first $500 of covered repair costs to your Honda; Northgate pays the covered amount above that, subject to the final claim decision. - If Pacific Shore (the other driver’s insurer)
CUSTOMERBut
AGENTI’ll keep it simple and stick
AGENTI hear you—there’s a lot to take in, and it’s all hitting you at once. Let me slow this down and keep it very simple, and you can stop me anywhere if something doesn’t sound right. First, just the key facts for you: - Your policy number: NG4471902. - The other driver’s license plate: 7KXR214. - Your claim number: NGM-904417. If any of those sound wrong, tell me and I’ll correct them. Now, in plain terms: 1) Tow truck & where your car is going. - A tow from Eastside Towing is on the way. - Estimated arrival: about 45 minutes. - Planned destination: a Northgate partner repair shop at 9th and Morrison in Portland. - When the driver arrives, you can say: - “Take it to the shop at 9th and Morrison,” or. - Give them a different shop you prefer. - Your policy covers towing up to 25 miles, so this tow should be covered. 2) Rental car. - Your policy includes rental coverage: $40 per day, up to 30 days, starting today. - That means: - If the rental is $40/day or less, it should be fully covered (within the 30‑day limit). - If it costs more than $40/day, you’d pay the
AGENTdifference. - Your adjuster, Priya Shah, will help you set up the rental and explain exactly how to do it. 3) What you might pay. - Your collision deductible is $500. - If your claim is handled through Northgate, you’d be responsible for up to $500 toward the covered repairs to your Honda. - Because you were rear‑ended, the other driver’s insurer (**Pacific Shore**) may pay for some or all of the damage instead. In that case, your out‑of‑pocket could be less or even nothing—your adjuster will sort that out for you. 4) Will your premium go up? - I can’t see or promise future rate changes from here. - Whether your premium changes will depend on: - Who is found at fault, and. - Northgate’s underwriting decision. - Your adjuster will review fault and then can tell you how that may affect your policy. For right now, the most important things for you are: - Wait somewhere safe for the tow truck. - Have your claim number NGM-904417 ready to show the tow driver or the repair shop. What’s the one thing you’re most worried about at this moment—the tow, the rental, the cost, or your premium? I’ll focus just on that next.