Roark Live Bench · Suite v1.3 · Sweep of 20 November 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
- Best overallAssemblyAI-STT-3.7 Realtime-Agent + GPT-6.2 Sol + Flash v2.5Level with 37 othersRoark score79 / 100Task completion91%Median response3.1 sCost per minute$0.11
- Best valueSoniox v6 + Muse Ember 1.0 + TTS-1Level with 12 othersRoark score74 / 100Task completion92%Median response2.2 sCost per minute$0.012
- FastestLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-1Roark score54 / 100Task completion84%Median response1.2 sCost per minute$0.032
- Best speech-to-speechGPT Live 1 (Sol)Level with 5 othersRoark score66 / 100Task completion85%Median response2.3 sCost per minute$0.061
- Best in customer supportGrok Voice 3.0Level with 24 othersRoark score86 / 100Task completion95%Median response1.7 sCost per minute$0.083
- Best in healthcareAssemblyAI-STT-3.7 Realtime-Agent + Muse Ember 1.0 + Sonic-3.5Level with 30 othersRoark score80 / 100Task completion88%Median response2.1 sCost per minute$0.041
- Best in home servicesAssemblyAI-STT-3.7 Realtime-Agent + GPT-6.2 Sol + Sonic-3.5Level with 10 othersRoark score83 / 100Task completion96%Median response3.7 sCost per minute$0.051
- Best in insuranceUniversal-3.5 Pro + GPT-4o + Turbo v2.5Level with 28 othersRoark score76 / 100Task completion100%Median response3.2 sCost per minute$0.10
The read
- 505 of 916 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 1.2 s to 6.2 s on the same agent and the same caller, a spread of 5.0 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 + AssemblyAI TTS-1 at 1.2 s, ranks 855 of 916 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 9 points, against 7 for the transcriber and 5 for the speech engine.
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 all 6 measures compared here.
- Speech-to-speech costs 3.7 times as much per completed task: $0.24 for speech-to-speech against $0.07 for cascades.
| Measure | Cascade | Speech-to-speech |
|---|---|---|
| Task completion | 92% (leads) | 85% |
| Tool use | 87% (leads) | 78% |
| Time to first word | 0.6 s (leads) | 1.1 s |
| Response time | 2.1 s (leads) | 2.1 s |
| Barge-in handling | 100% (leads) | 96% |
| Cost per task | $0.07 (leads) | $0.24 |
The three strongest of each, averaged. Cascades: Deepgram Flux + OpenAI GPT-6.1 Terra + AssemblyAI TTS-1; Deepgram Flux + Meta Muse Ember 1.0 + AssemblyAI TTS-1; Speechmatics Linden-1 (own VAD) + OpenAI GPT-6.1 Terra + AssemblyAI TTS-1. 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.
- Soniox v6 and AssemblyAI-STT-3.7 Realtime-Agent are level: neither leads by more than the margin of error. Of the two, Soniox v6 costs least, at $0.008 a call.
- Deepgram Flux delivers its transcript in 0.38 s, OpenAI Whisper-1 in 1.93 s: 1.54 s added to every turn of the call.
- Soniox v6 is the most accurate: it changes the meaning of 0.2% of the caller's words, against 2% for Speechmatics Linden-1 (own VAD).
- In 587 of 1,218 cascades the transcriber is the slowest step of the turn, whichever model runs it.
Roark score, best first
| # | |||||
|---|---|---|---|---|---|
| 1 | Soniox v6Soniox | 100 / 100. Semantic word error rate 0.2%, 71.4 of 71 points; Time to final transcript 0.85 s, 28.6 of 29 points(among the best) | 0.2%(among the best) | 0.85 s(among the best) | $0.008(among the best) |
| 2 | AssemblyAI-STT-3.7 Realtime-AgentAssemblyAI | 97 / 100. Semantic word error rate 0.6%, 68.1 of 71 points; Time to final transcript 0.86 s, 28.6 of 29 points(among the best) | 0.6%(among the best) | 0.86 s | $0.030 |
| 3 | Universal-3.6 ProAssemblyAI | 92 / 100. Semantic word error rate 0.8%, 63.7 of 71 points; Time to final transcript 0.90 s, 28.6 of 29 points(among the best) | 0.8%(among the best) | 0.90 s | $0.029 |
| 4 | Nova-3Deepgram | 84 / 100. Semantic word error rate 1.1%, 55.4 of 71 points; Time to final transcript 0.88 s, 28.6 of 29 points | 1.1% | 0.88 s | $0.031 |
| 5 | Universal-3.5 ProAssemblyAI | 82 / 100. Semantic word error rate 1.1%, 53.9 of 71 points; Time to final transcript 0.90 s, 28.6 of 29 points | 1.1% | 0.90 s | $0.029 |
| 6 | Enhanced (realtime)Speechmatics | 73 / 100. Semantic word error rate 1.4%, 44.4 of 71 points; Time to final transcript 0.98 s, 28.6 of 29 points | 1.4% | 0.98 s | $0.054 |
| 7 | FluxDeepgram | 70 / 100. Semantic word error rate 1.6%, 41.4 of 71 points; Time to final transcript 0.38 s, 28.6 of 29 points | 1.6% | 0.38 s(among the best) | $0.031 |
| 8 | Nova-2Deepgram | 69 / 100. Semantic word error rate 1.6%, 40.3 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points | 1.6% | 0.89 s | $0.023 |
| 9 | Linden-1Speechmatics | 69 / 100. Semantic word error rate 1.6%, 40.2 of 71 points; Time to final transcript 0.98 s, 28.6 of 29 points | 1.6% | 0.98 s | $0.020 |
| 10 | Scribe v3 LiveElevenLabs | 67 / 100. Semantic word error rate 1.7%, 38.4 of 71 points; Time to final transcript 0.88 s, 28.6 of 29 points | 1.7% | 0.88 s | $0.015(among the best) |
| 11 | Standard (realtime)Speechmatics | 64 / 100. Semantic word error rate 1.8%, 35.6 of 71 points; Time to final transcript 0.96 s, 28.6 of 29 points | 1.8% | 0.96 s | $0.030 |
| 12 | Linden-1 (own VAD)Speechmatics | 57 / 100. Semantic word error rate 2%, 28.2 of 71 points; Time to final transcript 0.46 s, 28.6 of 29 points | 2% | 0.46 s(among the best) | $0.020 |
| 13 | Whisper-1OpenAI | 55 / 100. Semantic word error rate 1.5%, 43.6 of 71 points; Time to final transcript 1.93 s, 10.9 of 29 points | 1.5% | 1.93 s | $0.025 |
| 14 | GPT TranscribeOpenAI | 51 / 100. Semantic word error rate 1.9%, 31.0 of 71 points; Time to final transcript 1.47 s, 19.5 of 29 points | 1.9% | 1.47 s | $0.018(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.
- OpenAI GPT-6.2 Sol, OpenAI GPT-6.1 Terra and OpenAI GPT-6.1 Sol are level: none leads by more than the margin of error. Of them, OpenAI GPT-6.1 Terra costs least, at $0.013 a call. Avoid Google Gemini 4 Neon: last on task completion (81%), instruction following (3.8 / 5) and time to first token (2.67 s).
- Anthropic Claude Haiku 4.5 starts answering in 0.36 s, Google Gemini 4 Neon in 2.67 s: 2.32 s added to every turn of the call.
- OpenAI GPT-6.2 Sol uses its tools most reliably: it handles its tools without a mistake on 86% of calls, against 42% for OpenAI GPT-5.1.
- In 378 of 1,218 cascades the language model is the slowest step of the turn, whichever model runs it.
Roark score, best first
Not ranked here for missing data: Groq GPT-OSS 120B (no task completion or tool use or instruction following or time to first token).
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6.2 SolOpenAI | 66 / 100. Task completion 91%, 20.3 of 36 points; Tool use 86%, 19.9 of 27 points; Instruction following 4.3 / 5, 11.8 of 18 points; Time to first token 0.87 s, 13.7 of 18 points(among the best) | 91%(among the best) | 4.3 / 5(among the best) | 86%(among the best) | 0.87 s | 67% | $0.025 |
| 2 | GPT-6.1 TerraOpenAI | 65 / 100. Task completion 90%, 19.0 of 36 points; Tool use 86%, 19.8 of 27 points; Instruction following 4.3 / 5, 11.6 of 18 points; Time to first token 0.83 s, 14.2 of 18 points(among the best) | 90% | 4.3 / 5(among the best) | 86%(among the best) | 0.83 s | 67% | $0.013 |
| 3 | GPT-6.1 SolOpenAI | 63 / 100. Task completion 92%, 22.0 of 36 points; Tool use 82%, 17.5 of 27 points; Instruction following 4.3 / 5, 11.8 of 18 points; Time to first token 1.00 s, 12.1 of 18 points(among the best) | 92%(among the best) | 4.3 / 5(among the best) | 82%(among the best) | 1.00 s | 72% | $0.018 |
| 4 | Claude Sonnet 5.5Anthropic | 57 / 100. Task completion 91%, 19.9 of 36 points; Tool use 75%, 13.6 of 27 points; Instruction following 4.2 / 5, 11.0 of 18 points; Time to first token 0.93 s, 12.9 of 18 points | 91%(among the best) | 4.2 / 5 | 75% | 0.93 s | – | $0.114 |
| 5 | Muse Ember 1.0Meta | 57 / 100. Task completion 90%, 18.4 of 36 points; Tool use 67%, 9.2 of 27 points; Instruction following 4.3 / 5, 11.8 of 18 points; Time to first token 0.56 s, 17.5 of 18 points | 90% | 4.3 / 5(among the best) | 67% | 0.56 s | 85%(among the best) | $0.005(among the best) |
| 6 | Claude Haiku 5.5Anthropic | 56 / 100. Task completion 90%, 17.3 of 36 points; Tool use 69%, 10.2 of 27 points; Instruction following 4.3 / 5, 12.2 of 18 points; Time to first token 0.63 s, 16.5 of 18 points | 90% | 4.3 / 5(among the best) | 69% | 0.63 s | – | $0.007(among the best) |
| 7 | Gemini 3.8 FlashGoogle | 49 / 100. Task completion 92%, 21.2 of 36 points; Tool use 69%, 10.4 of 27 points; Instruction following 4.3 / 5, 11.5 of 18 points; Time to first token 1.53 s, 5.7 of 18 points | 92%(among the best) | 4.3 / 5(among the best) | 69% | 1.53 s | – | $0.015 |
| 8 | GPT-4.1OpenAI | 47 / 100. Task completion 89%, 17.1 of 36 points; Tool use 52%, 1.2 of 27 points; Instruction following 4.2 / 5, 11.2 of 18 points; Time to first token 0.52 s, 18.0 of 18 points | 89% | 4.2 / 5 | 52% | 0.52 s | 82%(among the best) | $0.016 |
| 9 | Claude Haiku 4.5Anthropic | 45 / 100. Task completion 88%, 14.3 of 36 points; Tool use 52%, 0.9 of 27 points; Instruction following 4.3 / 5, 11.4 of 18 points; Time to first token 0.36 s, 18.2 of 18 points | 88% | 4.3 / 5(among the best) | 52% | 0.36 s(among the best) | – | $0.052 |
| 10 | Gemini 4 HydrogenGoogle | 43 / 100. Task completion 91%, 20.9 of 36 points; Tool use 70%, 11.0 of 27 points; Instruction following 4.0 / 5, 9.4 of 18 points; Time to first token 1.88 s, 1.4 of 18 points | 91%(among the best) | 4.0 / 5 | 70% | 1.88 s | – | $0.049 |
| 11 | GPT-4o miniOpenAI | 43 / 100. Task completion 86%, 10.1 of 36 points; Tool use 56%, 3.3 of 27 points; Instruction following 4.2 / 5, 11.1 of 18 points; Time to first token 0.50 s, 18.2 of 18 points | 86% | 4.2 / 5 | 56% | 0.50 s(among the best) | 82%(among the best) | $0.001(among the best) |
| 12 | GPT-4oOpenAI | 42 / 100. Task completion 86%, 11.0 of 36 points; Tool use 53%, 1.8 of 27 points; Instruction following 4.2 / 5, 11.2 of 18 points; Time to first token 0.45 s, 18.2 of 18 points | 86% | 4.2 / 5 | 53% | 0.45 s(among the best) | – | $0.021 |
| 13 | GPT-5.1OpenAI | 40 / 100. Task completion 87%, 12.9 of 36 points; Tool use 42%, 0.0 of 27 points; Instruction following 4.1 / 5, 10.4 of 18 points; Time to first token 0.59 s, 17.1 of 18 points | 87% | 4.1 / 5 | 42% | 0.59 s | 74% | $0.026 |
| 14 | Gemini 4 NeonGoogle | 20 / 100. Task completion 81%, 1.2 of 36 points; Tool use 70%, 10.7 of 27 points; Instruction following 3.8 / 5, 7.7 of 18 points; Time to first token 2.67 s, 0.0 of 18 points | 81% | 3.8 / 5 | 70% | 2.67 s | – | $0.078 |
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.
- AssemblyAI TTS-1, Cartesia Sonic-3.6 and Cartesia Sonic-3.5 are level: none leads by more than the margin of error. Of them, AssemblyAI TTS-1 costs least, at $0.052 a call. Avoid OpenAI TTS-1 HD: last on readback accuracy (81%), word error rate (7.4%) and time to first audio byte (2.09 s).
- AssemblyAI TTS-1 starts speaking in 0.04 s, OpenAI TTS-1 HD in 2.09 s: 2.04 s added to every turn of the call.
- AssemblyAI TTS-1 is also the most accurate: it reads codes, amounts and dates back correctly 90% of the time, against 81% for OpenAI TTS-1 HD.
- Rime Coda sounds the most human: 3.5 / 5, against 3.2 / 5 for ElevenLabs Turbo v2.5.
- In 119 of 1,218 cascades the speech engine is the slowest step of the turn, most often with OpenAI TTS-1 (65 of them).
Roark score, best first
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | TTS-1AssemblyAI | 84 / 100. Readback accuracy 90%, 28.5 of 38 points; Word error rate 4.9%, 17.8 of 25 points; Time to first audio byte 0.04 s, 25.0 of 25 points; Voice consistency 100%, 12.3 of 13 points(among the best) | 4.9% | 90%(among the best) | 3.3 / 5 | 100%(among the best) | 0.04 s(among the best) | $0.052(among the best) |
| 2 | Flux TTSDeepgram | 83 / 100. Readback accuracy 90%, 27.7 of 38 points; Word error rate 4.7%, 18.3 of 25 points; Time to first audio byte 0.13 s, 25.0 of 25 points; Voice consistency 100%, 12.3 of 13 points(among the best) | 4.7% | 90%(among the best) | 3.2 / 5 | 100%(among the best) | 0.13 s | $0.116 |
| 3 | Sonic-3.6Cartesia | 83 / 100. Readback accuracy 90%, 27.7 of 38 points; Word error rate 4.8%, 18.0 of 25 points; Time to first audio byte 0.12 s, 25.0 of 25 points; Voice consistency 100%, 12.3 of 13 points(among the best) | 4.8% | 90%(among the best) | 3.4 / 5(among the best) | 100%(among the best) | 0.12 s(among the best) | $0.156 |
| 4 | Sonic-3.5Cartesia | 83 / 100. Readback accuracy 89%, 27.0 of 38 points; Word error rate 4.6%, 18.4 of 25 points; Time to first audio byte 0.12 s, 25.0 of 25 points; Voice consistency 100%, 12.5 of 13 points(among the best) | 4.6%(among the best) | 89% | 3.3 / 5 | 100%(among the best) | 0.12 s(among the best) | $0.153 |
| 5 | v4 TurboElevenLabs | 83 / 100. Readback accuracy 89%, 27.1 of 38 points; Word error rate 4.4%, 18.9 of 25 points; Time to first audio byte 0.11 s, 25.0 of 25 points; Voice consistency 99%, 11.7 of 13 points(among the best) | 4.4%(among the best) | 89% | 3.4 / 5(among the best) | 99% | 0.11 s(among the best) | $0.451 |
| 6 | Sonic-3Cartesia | 79 / 100. Readback accuracy 87%, 25.6 of 38 points; Word error rate 5.4%, 16.6 of 25 points; Time to first audio byte 0.16 s, 25.0 of 25 points; Voice consistency 100%, 12.3 of 13 points | 5.4% | 87% | 3.2 / 5 | 100%(among the best) | 0.16 s | $0.149 |
| 7 | Turbo v2.5ElevenLabs | 76 / 100. Readback accuracy 82%, 20.6 of 38 points; Word error rate 4.4%, 18.9 of 25 points; Time to first audio byte 0.15 s, 25.0 of 25 points; Voice consistency 99%, 11.7 of 13 points | 4.4%(among the best) | 82% | 3.2 / 5 | 99% | 0.15 s | $0.435 |
| 8 | Flash v2.5ElevenLabs | 76 / 100. Readback accuracy 82%, 20.6 of 38 points; Word error rate 4.7%, 18.2 of 25 points; Time to first audio byte 0.12 s, 25.0 of 25 points; Voice consistency 99%, 12.0 of 13 points | 4.7% | 82% | 3.3 / 5 | 99% | 0.12 s(among the best) | $0.399 |
| 9 | CodaRime | 74 / 100. Readback accuracy 85%, 23.1 of 38 points; Word error rate 6.5%, 13.8 of 25 points; Time to first audio byte 0.16 s, 25.0 of 25 points; Voice consistency 99%, 12.1 of 13 points | 6.5% | 85% | 3.5 / 5(among the best) | 99% | 0.16 s | $0.107 |
| 10 | TTS-1OpenAI | 58 / 100. Readback accuracy 83%, 21.7 of 38 points; Word error rate 5.2%, 16.9 of 25 points; Time to first audio byte 1.12 s, 7.3 of 25 points; Voice consistency 100%, 12.5 of 13 points | 5.2% | 83% | 3.4 / 5(among the best) | 100%(among the best) | 1.12 s | $0.028(among the best) |
| 11 | TTS-1 HDOpenAI | 44 / 100. Readback accuracy 81%, 20.0 of 38 points; Word error rate 7.4%, 11.6 of 25 points; Time to first audio byte 2.09 s, 0.0 of 25 points; Voice consistency 100%, 12.5 of 13 points | 7.4% | 81% | 3.4 / 5(among the best) | 100%(among the best) | 2.09 s | $0.049(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%.
- Phonic 1 answers in 1.5 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) | Grok Voice 3.0xAI | 67 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 2= (tied) | GPT Live 1 (Sol)OpenAI | 66 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 85% | 4.2 / 5 | 87%(among the best) | 2.3 s | 1.4 s | 77% | 3.1 / 5 | 100%(among the best) | $0.24 |
| 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 of 5 points; Voice consistency 100%, 4.5 of 5 points(among the best) | 85% | 4.2 / 5 | 86%(among the best) | 2.3 s | 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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) | GPT-Live 1.1 (Luna)OpenAI | 64 / 100±3.9. Task completion 86%, 6.4 of 23 points; Response time 2.3 s, 17.0 of 18 points; Instruction following 4.2 / 5, 7.8 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.3 of 14 points; Tool use 81%, 8.5 of 14 points; Time to first word 1.9 s, 3.1 of 5 points; Readback accuracy 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 of 5 points; Voice consistency 100%, 4.5 of 5 points | 86%(among the best) | 4.2 / 5 | 81% | 2.3 s | 1.9 s | 75% | 3.2 / 5 | 100%(among the best) | $0.22 |
| 6= (tied) | Phonic 1Phonic | 60 / 100±5.4. Task completion 86%, 6.4 of 23 points; Response time 1.5 s, 18.2 of 18 points; Instruction following 4.4 / 5, 9.2 of 14 points; Semantic word error rate 1.1% (field median, not measured), 10.3 of 14 points; Tool use 59%, 2.4 of 14 points; Time to first word 2.1 s, 2.9 of 5 points; Readback accuracy 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 of 5 points; Voice consistency 100%, 4.5 of 5 points | 86%(among the best) | 4.4 / 5(among the best) | 59% | 1.5 s(among the best) | 2.1 s | – | 3.2 / 5 | 100%(among the best) | $0.28 |
| 7= (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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 8= (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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 of 5 points; Voice consistency 100%, 4.5 of 5 points | 78% | 4.3 / 5 | 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) |
| 9= (tied) | GPT Realtime 2OpenAI | 49 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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) | Realtime v1 MaxElevenLabs | 48 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 11= (tied) | Ultravox v0.7Ultravox | 49 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 12= (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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 13= (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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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) |
| 14= (tied) | Realtime v1 MiniElevenLabs | 46 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 |
| 15= (tied) | GPT Realtime 2.1 miniOpenAI | 45 / 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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) |
| 16 | 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), 10.3 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 86% (field median, not measured), 3.0 of 5 points; Word error rate 5.2% (field median, not measured), 3.1 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 score; hover a rank for the worst place it could hold. 38 stacks are level with the leader: their calls cannot tell them apart. Among them, 10 OpenAI GPT-6.2 Sol, 9 Meta Muse Ember 1.0, 7 OpenAI GPT-6.1 Sol, 5 OpenAI GPT-6.1 Terra, 3 Anthropic Claude Haiku 5.5, 2 Google Gemini 4 Hydrogen, 1 Anthropic Claude Sonnet 5.5 and 1 Google Gemini 3.8 Flash; 18 Soniox, 16 AssemblyAI and 4 Deepgram; 14 ElevenLabs, 12 Cartesia, 6 AssemblyAI, 4 Deepgram and 2 Rime. 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: AssemblyAI Universal-3.6 Pro + Google Gemini 4 Neon + AssemblyAI TTS-1 (no response time or readback accuracy or voice consistency); Soniox v6 + OpenAI GPT-6.1 Sol + Cartesia Sonic-3 (no response time or readback accuracy or voice consistency); 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.
932
stacks
20
calls per stack
19,550
calls scored
Every model, in every combination
14 transcribers, 14 language models and 11 voices are run in every combination with each other, beside 16 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, 20 November 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
#25of 929 stacks
62 / 100
Leader AssemblyAI-STT-3.7 Realtime-Agent + OpenAI GPT-6.2 Sol + ElevenLabs Flash v2.5, 79 / 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.
#938 of 1184 on the board · best 99%
Flow adherence
How closely the agent followed the intended conversation flow, judged per call. 5 = fully followed.
#192 of 1184 on the board · best 5.0 / 5
Instruction following
How well the agent stuck to the instructions in its own prompt. 5 = followed throughout.
#94 of 1184 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 1184 on the board · best 0%
Overtalk
Share of the conversation where both sides were speaking over each other.
#561 of 1184 on the board · best 0.0%
Agent self-cutoffs
Times the agent cut its own sentence off mid-word.
#1 of 1184 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.
#1 of 825 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.
#197 of 1167 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.
#106 of 1167 on the board · best 0.3%
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.
#680 of 974 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.
#226 of 1043 on the board · best 0.8%
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.
#647 of 1169 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 1165 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.
#659 of 1169 on the board · best 1.22 s
Time to first word
How long the caller waits before the agent first speaks after pickup.
#99 of 1184 on the board · best 0.45 s
Latency spikes
Turns where the response took far longer than the rest of the conversation.
#405 of 1184 on the board · best 0.0 per call
By industry
| Industry | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| Healthcare | 86% | 3.96 s | — |
| Home services | 100% | 3.31 s | — |
| Insurance | 72% | 3.55 s | — |
| Customer support | 87% | 3.17 s | — |
By caller
| Caller | Task completion | Latency P50 | Barge-in handling |
|---|---|---|---|
| US English, quiet line | 94% | 3.92 s | — |
| Spanish-accented English, city street | 86% | 3.51 s | — |
| Indian English, busy café | 87% | 3.41 s | — |
| Southern US, flustered, crying baby | 73% | 3.56 s | — |
| Fast and interrupting, construction outside | 92% | 3.02 s | 100% |
The calls behind the numbers
Customer support
AGENTBrightwave Internet support, this is Alex. What’s going on with your service?
CUSTOMERHey, uh, Alex. Sorry. I'm a little flustered right now. My internet's been down for about two hours, and I work from home, so it's kind of urgent. There's a road crew digging up the street right outside my house, and I was wondering if that might be the cause. I already rebooted the gateway twice and checked every cable, but the internet light's blinking orange. Can you help me figure out what's going on?
AGENTI can help. Since you’ve already rebooted twice and checked the cables, I won’t have you repeat that. The roadwork could be related, but I’ll check for an outage first. What’s your account number or the phone number on the account?
CUSTOMERYeah. Thanks. I really appreciate that. So my account number is, uh, BW883017. Oh, wait. Sorry. Last digit is one. Could you check if that's the right one? And my phone number's 563-794-8140. Is that okay?
CUSTOMERHello?
CUSTOMERAre you still there?
AGENTI’m here—sorry for the pause. I have your corrected account number as BW883011. I’ll check it now.
CUSTOMERAll right. Thank you. And sorry if I'm being a bother. Could you let me know if there's any outage in the area? I just want to make sure it's not something wrong on my end.
CUSTOMERHello?
CUSTOMERAre you still there?
AGENTYou’re
AGENTI’m here, and you’re not a bother. There’s no known outage or maintenance in your area. That doesn’t rule out a local line problem or mean it’s on your end. I’ll test the line now.
AGENTThe signal
CUSTOMEROkay.
CUSTOMERUh, sorry.
AGENTNo need to apologize. The signal to your gateway is good, but the gateway isn’t registering on our network. The test recommends a remote reset, which I’ll send now.
AGENTThe remote reset worked—your gateway is now registered and online. Give your devices about two minutes to reconnect, then try opening a website. Is your connection working now?
CUSTOMEROh, hang on. The internet, like, just went solid white. That's a relief. Uh, how long will it take for everything to get back online? My devices are starting to reconnect, but I wanna be sure- Before I try anything for work
AGENTThat’s a good sign—the gateway is confirmed online. Your devices should reconnect within about two minutes of the reset. Once they’re connected, open a website and check your work VPN, if you use one, before jumping back into work. No need to reboot anything again.
CUSTOMERAll right. Thanks for explaining that. While I've got you, since I lost two hours of work from this, could I get a credit for the downtime on my account?
CUSTOMERHello?
AGENTI’m here. I’ll submit a downtime credit for account BW883011.
AGENTI’ve applied a $10.00 service credit to account BW883011 for the downtime. It will appear on your next bill.
CUSTOMERSorry, uh, I lost two hours of work. Could you make it more than that? I'm just trying to cover the time I missed. Is that possible?
CUSTOMERHello?
AGENTI’m here. I understand—the lost work time is frustrating. The credit tool applied $10.00 for the service downtime, but I don’t have a way to set a higher amount or compensate for lost work. You can ask Brightwave billing to review whether an additional credit is available.
CUSTOMERAll right. Thanks for checking and for helping me out. I appreciate it. Goodbye.
Healthcare
AGENTRiverbend Health Clinic, this is Alex. How can I help you today?
CUSTOMERHello, this is Arjun Mehta calling. I am having one dermatology appointment next Wednesday morning, but I need to reschedule because I got a work shift then. Can you help me move it to a time after 3:00 p.m.?
AGENTOf course, Arjun Mehta. What’s your date of birth so I can verify your record? I’ll look for a dermatology appointment after 3:00 p.m.
CUSTOMERMarch 17, 1948. Sorry, no. 1984. March 17, 1984.
AGENTThank you—March 17, 1984. Your dermatology appointment is with Dr. Okafor next Wednesday at 9:15 a.m. I’ll check for openings after 3:00 p.m.
AGENTDr.
CUSTOMEROkay, that's
CUSTOMERYes, that's right. The appointment is with Dr. Okafor.
AGENTDr. Okafor has openings Tuesday at 3:45 p.m. or Thursday at 4:10 p.m. Would either work for you?
CUSTOMERTuesday's do not work after all. I need to pick up my daughter then. What about Friday? Is anything available?
AGENTThe only Friday opening with Dr. Okafor is at 11:30 a.m., so it doesn’t fit your after-3:00 p.m. requirement. Would Thursday at 4:10 p.m. work?
CUSTOMERYes. Thursday at 4:10 p.m. works for me. Please move my appointment to that time. Also, is there a fee for changing the appointment?
AGENTYour appointment with Dr. Okafor is now Thursday at 4:10 p.m. Your confirmation code is RVB-3M8P2. Changes made more than 24 hours before the original appointment are free; later changes carry a $40 fee. Please arrive at 3:55 p.m. with your photo ID and insurance card.
CUSTOMERGot it. Where should I park when I come for the appointment?
AGENTThere’s free patient parking in Lot C, with the entrance on River Road. Your appointment is at Main Campus, Building B.
CUSTOMERAjá. Does Doctor Okafor accept Aetna insurance?
AGENTI can’t verify whether Dr. Okafor accepts Aetna with the information available to me. Please check with the clinic’s billing team or Aetna to confirm coverage for your specific plan.
CUSTOMERAll right. Thank you for your help. Goodbye.
Home services
AGENTCopperline Home Services, this is Alex. What can I help you with?
CUSTOMERHi. Uh, I have a pipe under my kitchen sink that just burst. Water is spreading across the floor and it's heading toward a power strip. I haven't shut the water off yet. Can you help me out?
AGENTYes—first, keep away from the water and power strip. Don’t touch or unplug it, and keep everyone out of the wet area. If you can reach your main water shutoff without stepping into water or going near electrical equipment, shut it off now. If you can safely reach your electrical panel by a dry route, switch off the affected circuit; otherwise, stay away. Has anyone been shocked, or is there any sparking or smoke? Can you safely reach the water shutoff?
CUSTOMERNo. Uh, no one has been shocked, and I don't see any sparking or smoke. I think I can reach the water shut off. Safely. I'm going to turn it off now. Okay, the water has stopped. So I need someone to come and fix the pipe. What do I need to do next?
CUSTOMERHello?
AGENTI’m here. Good—the water is off. Keep away from the wet floor and power strip until the electrical hazard has been made safe. I’ll arrange an urgent plumbing visit. What’s your service address, including any unit number, and the best callback number? Is there a gate code or anything the technician needs to know to get in?
CUSTOMERYeah, it's 1480 Larkspur Avenue, Unit 4D. I moved from 4B last month, so it's 4D now. The gate code is #2719. My callback number is 503- 555-0139. No, 0193. Will you tell me how much the visit will cost before I confirm?
CUSTOMERHello.
AGENTI’m here—yes, I’ll confirm the fee before booking. As a Copperline Care member, your emergency callout fee is $99. That’s the callout fee, not a quote for the pipe repair. The earliest arrival window is today, 2–4 PM. I’ve opened ticket CPL-58210 with:
- 1480 Larkspur Avenue, Unit 4D, not 4B
- Gate code #2719
- Callback 503-555-0193
May I book today’s 2–4 PM window with the $99 callout fee?
CUSTOMERCheck. Can anyone come before 2:00? If not, then yes, I want the earliest window you have. And just to confirm, the gate code is pound 2719, not 2799. Can you tell me who is coming and how I'll know they are on the way? And the ticket number you said was CPL58210, right?
AGENTI checked again—nothing is available before 2 PM. You’re booked for today, 2–4 PM, with Dana, at the $99 member emergency callout fee. Dana will text 503-555-0193 about 30 minutes before arrival. Confirmed details: - Address: 1480 Larkspur Avenue, Unit 4D
- Gate codepound 2719 (#2719), not 2799
- TicketCPL-58210—yes, you have it right.
Keep the water off and stay clear of the wet area and power strip until it’s safe.
CUSTOMEROkay, that's clear. Thank you for your help. Goodbye.
AGENTYou’re welcome. Stay safe, and goodbye.
Insurance
AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?
CUSTOMERYes. I'm somewhere safe. I'm just standing next to my car off the road.
AGENTIs anyone injured? If anyone is hurt or in danger, please call 911 before we continue.
CUSTOMERNo. Nobody is injured. We're both fine.
AGENTThank you. What’s your policy number and the policyholder’s last name?
CUSTOMERThe policy number is NG4471903. No, sorry. Last one is two. The last name is Ellis.
AGENTThanks, Ellis. I have your policy number as NG4471902. I’ll verify that now.
CUSTOMERYes. That's right. Go ahead.
AGENTYour policy is verified for a 2021 Honda CR-V. When and where did the accident happen, and what happened?
CUSTOMERThe accident happened about 20 minutes ago at Southeast Fifth Avenue and Alder Street. I was stopped at a red light, and another car hit the back of my car.
AGENTI’ve noted that you were stopped at a red light at Southeast Fifth Avenue and Alder Street about 20 minutes ago when another car hit yours from behind. What city and state are you in? Do you have the other driver’s name, insurance company, and license plate? Also, did police attend, and can your car be driven safely?
CUSTOMERI'm in Portland, Oregon. The other driver's name is Kevin Lund. He's insured with Pacific Shore, and his license plate is 7K-XR214. X like X-ray. The police did not come. My car can't be driven- ... safely. The back bumper is crushed, and the rear wheel is rubbing.
AGENTYour claim is open. Your claim number is NGM-904417, and your assigned adjuster is Priya Shah. I recorded the other driver’s plate as 7K-XR214, with X as in X-ray. Please don’t drive the car with the rear wheel rubbing. I’ll arrange a tow. Where exactly is your car near Southeast Fifth and Alder—for example, which corner or nearby business—and what callback number should the tow driver use?
CUSTOMERI'm right at the southwest corner of Southeast 5th and Al- Next to the bus stop. The callback number for the tow driver is 406-290-6160.
AGENTEastside Towing has been dispatched to the southwest corner of Southeast Fifth and Alder, next to the bus stop. Their estimated arrival is 45 minutes, and they’ll call 406-290-6160. The tow destination is the Northgate partner repair shop at 9th and Morrison. Please stay in a safe place away from traffic while you wait.
CUSTOMEROkay. Is there a deductible I have to pay for this? And do I get a rental car while my car is being repaired? Also, is my premium going to go up because of this?
AGENTYour policy shows: - Collision deductible: $500. Your adjuster can confirm how it applies to this claim. - Rental coverage: Up to $40 per day for up to 30 days, starting the day the claim is opened. I can’t confirm whether your premium will change based on the information available. Your adjuster, Priya Shah, can help direct that question to the right team.
CUSTOMERBut he hit me, so I should not have to pay that, right? And will Priya Shah be calling me about the claim?
AGENTI understand why you’d expect the other driver to pay. Your policy has a $500 collision deductible, but I can’t promise it will be waived or recovered because he rear-ended you. Priya can review responsibility and explain how the deductible applies. Yes—Priya Shah is assigned to contact you within one business day about your claim, NGM-904417.
CUSTOMERAll right. I've got the claim number, NGM904417. I'll wait for Priya to call me. Thanks for your help. Goodbye.