Skip to content

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.

662 stacks published14,070 scored callsPublished at 20 scored calls

Our picks

Roark scoreTask completionMedian responseCost per minute

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.
CascadeSpeech-to-speech
Task completionTool useTime to first wordResponse timeBarge-in handlingCost per task255075100
MeasureCascadeSpeech-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 s2.1 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

90Universal-3.6 Pro84Universal-3.5 Pro82Nova-382Enhanced (realtime)79Linden-171Nova-266Standard (realtime)56Whisper-152Linden-1 (own VAD)
Every transcriber on every figure, sorted by roark score. Column headers sort the table.
#
1Universal-3.6 ProAssemblyAI90 / 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
2Universal-3.5 ProAssemblyAI84 / 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
3Nova-3Deepgram82 / 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
4Enhanced (realtime)Speechmatics82 / 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
5Linden-1Speechmatics79 / 100. Semantic word error rate 1.3%, 50.0 of 71 points; Time to final transcript 0.99 s, 28.6 of 29 points1.3%0.99 s$0.020(among the best)
6Nova-2Deepgram71 / 100. Semantic word error rate 1.5%, 42.2 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points1.5%0.89 s(among the best)$0.023(among the best)
7Standard (realtime)Speechmatics66 / 100. Semantic word error rate 1.7%, 37.5 of 71 points; Time to final transcript 0.96 s, 28.6 of 29 points1.7%0.96 s$0.031
8Whisper-1OpenAI56 / 100. Semantic word error rate 1.4%, 45.5 of 71 points; Time to final transcript 1.93 s, 10.9 of 29 points1.4%1.93 s$0.025
9Linden-1 (own VAD)Speechmatics52 / 100. Semantic word error rate 2.2%, 23.1 of 71 points; Time to final transcript 0.49 s, 28.6 of 29 points2.2%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.

Cost against qualitySemantic word error rate · Cost per call
$0$0.02$0.04$0.06$0.080.5%1%1.5%2%2.5%Cheap and accurateAccurate, pricierCheap, less accuratePricier and less accurateLinden-1 (own VAD)Linden-1Nova-2Whisper-1Universal-3.6 ProUniversal-3.5 ProStandard (realtime)Nova-3Enhanced (realtime)Cost per call · ← betterSemantic word error rate · better →
Speed against qualitySemantic word error rate · Time to final transcript
0 s1 s2 s3 s0.5%1%1.5%2%2.5%Fast and accurateAccurate, slowerFast, less accurateSlower and less accurateLinden-1 (own VAD)Nova-3Nova-2Universal-3.6 ProUniversal-3.5 ProStandard (realtime)Enhanced (realtime)Linden-1Whisper-1Time to final transcript · ← betterSemantic word error rate · better →

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

66GPT-6.1 Sol61Claude Sonnet 5.551Gemini 3.8 Flash50GPT-4.149Claude Haiku 4.547GPT-4o mini44GPT-5.144GPT-4o
Every language model on every figure, sorted by roark score. Column headers sort the table.
#
1GPT-6.1 SolOpenAI66 / 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 s73%$0.019(among the best)
2Claude Sonnet 5.5Anthropic61 / 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
3Gemini 3.8 FlashGoogle51 / 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)
4GPT-4.1OpenAI50 / 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 points92%4.2 / 5(among the best)48%0.52 s82%(among the best)$0.022
5Claude Haiku 4.5Anthropic49 / 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 points91%4.2 / 5(among the best)47%0.36 s(among the best)–$0.062
6GPT-4o miniOpenAI47 / 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 points90%4.2 / 5(among the best)51%0.45 s(among the best)81%(among the best)$0.002(among the best)
7GPT-5.1OpenAI44 / 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 points89%4.1 / 538%0.59 s74%(among the best)$0.031
8GPT-4oOpenAI44 / 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 points88%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.

Cost against qualityTask completion · Cost per call
$0$0.05$0.1$0.15$0.285%90%95%100%Cheap and reliableReliable, pricierCheap, less reliablePricier and less reliableGPT-4o miniGemini 3.8 FlashGPT-6.1 SolGPT-4.1GPT-4oGPT-5.1Claude Haiku 4.5Claude Sonnet 5.5Cost per call · ← betterTask completion · better →
Speed against qualityTask completion · Time to first token
0 s0.5 s1 s1.5 s2 s85%90%95%100%Fast and reliableReliable, slowerFast, less reliableSlower and less reliableClaude Haiku 4.5GPT-4o miniGPT-4oGPT-4.1GPT-5.1Claude Sonnet 5.5GPT-6.1 SolGemini 3.8 FlashTime to first token · ← betterTask completion · better →

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

77Sonic-3.677v4 Turbo77Sonic-3.573Sonic-371Turbo v2.570Flash v2.568Coda55TTS-139TTS-1 HD
Every speech engine on every figure, sorted by roark score. Column headers sort the table.
#
1Sonic-3.6Cartesia77 / 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
2v4 TurboElevenLabs77 / 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 / 598%0.11 s(among the best)$0.470
3Sonic-3.5Cartesia77 / 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
4Sonic-3Cartesia73 / 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 points6.9%85%3.4 / 599%0.17 s$0.165
5Turbo v2.5ElevenLabs71 / 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 points5.7%(among the best)80%3.4 / 598%0.15 s$0.451
6Flash v2.5ElevenLabs70 / 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 points6%(among the best)80%3.4 / 598%0.12 s(among the best)$0.437
7CodaRime68 / 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 points8.1%83%3.7 / 5(among the best)99%0.18 s$0.116(among the best)
8TTS-1OpenAI55 / 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 points6.1%82%3.6 / 5100%(among the best)1.13 s$0.036(among the best)
9TTS-1 HDOpenAI39 / 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 points8.7%80%3.6 / 5100%(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.

Cost against qualityReadback accuracy · Cost per call
$0$0.2$0.4$0.675%80%85%90%Cheap and accurateAccurate, pricierCheap, less accuratePricier and less accurateTTS-1TTS-1 HDCodaSonic-3Sonic-3.5Sonic-3.6Flash v2.5Turbo v2.5v4 TurboCost per call · ← betterReadback accuracy · better →
Speed against qualityReadback accuracy · Time to first audio byte
0 s1 s2 s3 s75%80%85%90%Fast and accurateAccurate, slowerFast, less accurateSlower and less accuratev4 TurboSonic-3.6Flash v2.5Sonic-3.5Turbo v2.5Sonic-3CodaTTS-1TTS-1 HDTime to first audio byte · ← betterReadback accuracy · better →

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

65GPT Live 1 (Sol)66Grok Voice 3.065GPT Live 1 (Luna)65Grok Voice Think Fast 2.053Gemini 3.8 Live53GPT Realtime 2.148Ultravox v0.748Gemini 3.1 Flash Live (preview)48GPT Realtime 246Nova 2 Sonic47Realtime v1 Max45Realtime v1 Mini44GPT Realtime 2.1 mini36Gemini 2.5 Flash native audio

Whiskers show the 95% interval. The shaded models tie for first: these calls cannot separate them.

Every speech-to-speech model on every figure, sorted by roark score. Column headers sort the table.
#
1= (tied)GPT Live 1 (Sol)OpenAI65 / 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 / 587%(among the best)2.3 s(among the best)1.4 s77%3.1 / 5100%(among the best)$0.24
2= (tied)Grok Voice 3.0xAI66 / 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 s94%3.0 / 5100%(among the best)$0.26
3= (tied)GPT Live 1 (Luna)OpenAI65 / 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 / 586%(among the best)2.3 s(among the best)1.5 s93%3.1 / 5100%(among the best)$0.21
4= (tied)Grok Voice Think Fast 2.0xAI65 / 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 s100%(among the best)3.6 / 5100%(among the best)$0.26
5= (tied)Gemini 3.8 LiveGoogle53 / 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 points78%4.3 / 5(among the best)69%2.8 s1.6 s100%(among the best)4.1 / 5(among the best)100%(among the best)$0.14(among the best)
6= (tied)GPT Realtime 2.1OpenAI53 / 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 points78%3.8 / 575%2.6 s0.8 s(among the best)83%3.9 / 5(among the best)99%$0.43
7= (tied)Ultravox v0.7Ultravox48 / 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 points79%3.7 / 561%2.4 s1.6 s94%3.6 / 5100%(among the best)$0.22
8= (tied)Gemini 3.1 Flash Live (preview)Google48 / 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 points80%4.2 / 538%2.7 s1.5 s100%(among the best)4.0 / 5(among the best)100%(among the best)$0.15
9= (tied)GPT Realtime 2OpenAI48 / 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 points76%3.8 / 564%2.9 s0.8 s(among the best)91%3.7 / 5100%(among the best)$0.69
10= (tied)Nova 2 SonicAmazon46 / 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 points69%4.2 / 548%3.0 s1.5 s100%(among the best)3.5 / 5100%(among the best)$0.06(among the best)
11= (tied)Realtime v1 MaxElevenLabs47 / 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 points56%4.0 / 540%2.6 s0.7 s(among the best)76%3.9 / 5(among the best)100%(among the best)$0.65
12= (tied)Realtime v1 MiniElevenLabs45 / 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 points42%3.8 / 534%2.7 s0.7 s(among the best)94%3.4 / 5100%(among the best)$0.64
13= (tied)GPT Realtime 2.1 miniOpenAI44 / 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 points70%3.6 / 551%2.8 s1.0 s84%3.9 / 5(among the best)100%(among the best)$0.14(among the best)
14Gemini 2.5 Flash native audioGoogle36 / 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 points77%3.9 / 540%4.2 s3.4 s100%(among the best)3.7 / 5100%(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.

Cost against qualityTask completion · Cost per task
$0$0.5$120%40%60%80%100%Cheap and reliableReliable, pricierCheap, less reliablePricier and less reliableNova 2 SonicGemini 2.5 Flash native audioGPT Realtime 2.1 miniGemini 3.8 LiveGemini 3.1 Flash Live (preview)GPT Live 1 (Luna)Ultravox v0.7GPT Live 1 (Sol)Grok Voice 3.0Grok Voice Think Fast 2.0GPT Realtime 2.1Realtime v1 MiniRealtime v1 MaxGPT Realtime 2Cost per task · ← betterTask completion · better →
Speed against qualityTask completion · Response time
1 s2 s3 s4 s5 s20%40%60%80%100%Fast and reliableReliable, slowerFast, less reliableSlower and less reliableGrok Voice 3.0Grok Voice Think Fast 2.0GPT Live 1 (Luna)GPT Live 1 (Sol)Ultravox v0.7Realtime v1 MaxGPT Realtime 2.1Gemini 3.1 Flash Live (preview)Realtime v1 MiniGemini 3.8 LiveGPT Realtime 2.1 miniGPT Realtime 2Nova 2 SonicGemini 2.5 Flash native audioResponse time · ← betterTask completion · better →

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).

Loading every stack

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.

01

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

02

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.

03

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

FigureWeightFull marksZero
Task completionscenario_adherence_check23%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_time18%2.0 s6.0 s
In real calls, replying faster than 2 s made no measurable difference, and calls went worse past 3 s.
Instruction followinginstruction_follow14%5.0 / 53.0 / 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_rate14%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_use14%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_word5%1.0 s4.0 s
A greeting within 1 s sounds natural; 4 s of silence sounds like a dead line.
Readback accuracytts_entity_readback_accuracy5%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_rate5%2%12%
At 2% the voice says what it was given; at 12% one word in eight comes out wrong.
Voice consistencyvoice_consistency_check5%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

FigureWeightFull marksZero
Task completionscenario_adherence_check36%100%80%
An agent that completes fewer than four calls in five is not fit to deploy.
Tool usetool_use27%100%50%
An agent that mishandles its tools on half its calls cannot be trusted with the systems it runs on.
Instruction followinginstruction_follow18%5.0 / 53.0 / 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_token18%0.50 s2.00 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

FigureWeightFull marksZero
Semantic word error ratestt_semantic_word_error_rate71%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_transcript29%1.00 s2.50 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

FigureWeightFull marksZero
Readback accuracytts_entity_readback_accuracy38%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_rate25%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_byte25%0.20 s1.50 s
Under 0.2 s is not noticeable; 1.5 s is half the point where calls go worse.
Voice consistencyvoice_consistency_check13%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.

04

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.

Transcript, verbatim.

Read the full methodology

See how your own agent compares

Roark runs the same kind of calls against your agent, on your scenarios, with the same scoring.

Test your agent

Nova-2 + GPT-5.1 + v4 Turbo

Cascade

Rank on Roark score

#257of 655 stacks

45 / 100

Leader AssemblyAI Universal-3.5 Pro + OpenAI GPT-6.1 Sol + ElevenLabs Turbo v2.5, 70 / 100

Where it ranks

Roark score with price42 / 100543 of 655LeaderUniversal-3.5 Pro + GPT-6.1 Sol + Coda71 / 100
Task completion89%484 of 656LeaderLinden-1 (own VAD) + GPT-6.1 Sol + TTS-1 HD99%
Tool use37%592 of 632LeaderUniversal-3.6 Pro + GPT-6.1 Sol + TTS-1 HD90%
Time to first word0.7 s64 of 656LeaderUniversal-3.5 Pro + GPT-6.1 Sol + Flash v2.50.5 s
Response time3.5 s261 of 655LeaderGrok Voice 3.01.9 s
Barge-in handlingn/aLeaderWhisper-1 + GPT-6.1 Sol + v4 Turbo100%
Turn budget3.5 s229 of 596LeaderLinden-1 (own VAD) + GPT-4o mini + Flash v2.52.1 s
Cost per task$1.16641 of 655LeaderLinden-1 + GPT-4o mini + TTS-1$0.05

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.

20 calls
89%Completed

#489 of 662 on the board · best 99%

Flow adherence

How closely the agent followed the intended conversation flow, judged per call. 5 = fully followed.

20 calls
4.3/ 5avg

med

4

#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.

20 calls
4/ 5avg

med

4

#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.

20 calls
0.00%avg

p50

0.00%

p95

0.00%

#1 of 662 on the board · best 0%

Overtalk

Share of the conversation where both sides were speaking over each other.

20 calls
2.38%avg

p50

0.59%

p95

8.34%

#439 of 662 on the board · best 0.0%

Agent self-cutoffs

Times the agent cut its own sentence off mid-word.

20 calls
0.00per callavg

p50

0.00 per call

p95

0.00 per call

#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.

29 calls
0.69per callavg

p50

1.00 per call

p95

2.00 per call

#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.

20 calls
3.83%avg

p50

0.56%

p95

12.6%

#619 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.

20 calls
5.87%avg

p50

2.43%

p95

13.9%

#615 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.

19 calls
72.3%avg

p50

82.0%

p95

100%

#528 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.

20 calls
12.4%avg

p50

9.84%

p95

40.1%

#623 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.

20 calls
3.5/ 5avg

med

4

#330 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.

20 calls

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.

128 turns
3.56savg

p50

3.49s

p95

6.28s

#263 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.

20 calls
0.74savg

p50

0.71s

p95

1.40s

#64 of 662 on the board · best 0.45 s

Latency spikes

Turns where the response took far longer than the rest of the conversation.

20 calls
4.65per callavg

p50

4.50 per call

p95

9.00 per call

#205 of 662 on the board · best 0.0 per call

By industry

IndustryTask completionLatency P50Barge-in handling
Healthcare95%3.36 s—
Home services84%3.74 s—
Insurance92%3.24 s—
Customer support85%3.55 s—

By caller

CallerTask completionLatency P50Barge-in handling
US English, quiet line98%4.04 s—
Spanish-accented English, city street94%3.23 s—
Indian English, busy café98%3.26 s—
Southern US, flustered, crying baby96%3.85 s—
Fast and interrupting, construction outside61%3.41 s—

The calls behind the numbers

Customer support

AGENTBrightwave Internet support, this is Alex. What’s going on with your service?

CUSTOMERHi, Alex. Oh, yes. So I'm calling because my Internet has been down for about two hours now. I work from home, so it's, you know, a problem. There's a road crew outside my building digging up the street. So I'm wondering if that's the cause. I already rebooted the gateway twice, checked all the cables, The Internet light is blinking orange. You help me figure out what's going on?

AGENTI can help with that. Since you’ve already rebooted and checked cables, we won’t redo those. First I need to pull up your account and check for any area issue.

CUSTOMEROkay. My account number is BW883017. Sorry. Last digit is one. Can you check if there's any outage because of the road work?

AGENTThanks for that account number — I’ve got you under BW883017, Brightwave 500 at the address we have on file. Here’s what I’m seeing and what I did:

1. No area outage - I checked your locationthere’s no reported outage or maintenance in your area, so whatever the road crew is doing hasn’t been logged as a network outage on our side. 2. Line test & gateway status - The line from our network to your gateway looks good. - Your gateway was showing as “not registered” (basically, it wasn’t checking in with our

CUSTOMERUh-huh. Okay. What can we do next? Because the Internet light is still blinking orange. And I need to get back online for work.

AGENTI just sent a remote reset to your gateway, and it’s now showing as registered and online on my side. Let’s do two quick checks:

1. Look at the gateway lights - After a reset, it can take up to 2 minutes to settle. - The Internet light should go from blinking orange to solid white/blue (or solid green, depending on your BX7 model). - Can you tell me what the Internet light is doing right now? 2. Test on a device - On a computer or phone that’s connected to your Wi‑Fi (or directly by Ethernet), try- Opening a couple of different websites, or - Running a quick speed test at speedtest.net. If the light is still blinking orange after

CUSTOMEROh, the light just went solid white now. How long until my device will reconnect?

AGENTSolid white is what we want—that means the connection is back up. Most devices will reconnect within 1–2 minutes on their own. If you want to speed it up:

- On Wi‑Fi devicestoggle Wi‑Fi

CUSTOMEROkay. Got it. So, I'll wait a minute or two for everything to reconnect. Since I lost... Two hours of work, can you give me a credit for the downtime?

AGENTI can do that for you. I’ve applied a $10.00 credit to your account BW883017 for approximately two hours of service interruption. You’ll see it on your next bill. Once your devices finish reconnecting, if anything still seems off—like very slow speeds or frequent drops—tell me what you’re seeing on one wired device (if you have one) and I’ll help you fine-tune it.

CUSTOMERThanks. But I lost two hours of work. You make it more than that?

AGENTI understand why that feels frustrating, especially when you’re working from home. I’ve already applied the $10.00 credit to account BW883017, and that’s the standard adjustment we’re allowed to issue for a short interruption like this. I don’t have the ability to increase it beyond that amount. What I can do is:

- Make sure the connection is fully stable now so you’re not losing more time. - If you have any work tools (VPN, video calls, remote desktop) that seem sluggish or keep dropping, I can help optimize things (like checking for congestion, band steering between 2.4/5 GHz,

CUSTOMERNo. That's alright. I just wanted to get back online and get a fair credit. Thanks for your help, Alex. Goodbye.

AGENTYou’re welcome. The $10.00 credit is on your account BW883017 and your line looks healthy now, so you should be good to go. If anything acts up again, just reach out and we’ll dig into it right away. Take care.

Healthcare

AGENTRiverbend Health Clinic, this is Alex. How can I help you today?

CUSTOMERHi. This is Arjun Mehta. I am having a dermatology appointment next Wednesday morning, but I need to move it because I have a work shift then. Can you help me reschedule? I can only come after 3PM.

AGENTI can help with that. To look up your appointment, I need to verify your account first. Please tell me:

- Your full

CUSTOMERSure. My full name is Arjun Mehta.

AGENTThanks, Arjun. I also need your date of birth to look up your appointment. Could you please provide your date of birth (YYYY-MM-DD)?

CUSTOMER03/17/1948. Sorry. No. 1984. March 17. 1984.

AGENTI’ve found your appointment, Arjun:

- Current visitDermatology with Dr. Okafor

- Whennext Wednesday at 09:15 AM

- WhereMain Campus, Building B

You mentioned you can only come after 300 PM. Looking at Dr. Okafor’s schedule starting next Wednesday, the first options that fit “after 3 PM” are:

- Thursday at 410 PM

- (Also available, though not after

CUSTOMERThursday at 04:10PM works for me. Please book that one.

AGENTYou’re all set, Arjun. I’ve moved your dermatology appointment with Dr. Okafor from next Wednesday at 09:15 AM to:

- New dayThursday - New time: 4:10 PM - Location: Main Campus, Building B - Confirmation code: RVB-3M8P2

Please arrive 15 minutes early with a photo ID and your insurance card. Anything else I can help you with today?

CUSTOMERYes. A few quick questions. First, is there any fee for changing the appointment?

AGENTThere’s no fee for the change you just made. Our policy is: changes made more than 24 hours before the appointment are free; changes within 24 hours of the appointment have a $40 fee. You rescheduled well in advance, so there’s no charge.

CUSTOMEROkay. That's clear. Next, where should I park for Building B?

AGENTFor Building B, you can use our free patient parking in Lot C; the entrance is on River Road. From Lot C, follow the signs to Main Campus, Building B—it's a short walk from the lot.

CUSTOMERGot it. One more question. Does doctor Okafur accept Aetna insurance?

AGENTI don’t have access to provider-specific insurance panel details, so I can’t safely confirm that for Dr. Okafor. To get an accurate answer, you have two good options:

1) Call our front desk and ask if Dr. Okafor is currently in-network for Aetna for dermatology visits. 2) Check with Aetna (website or member services) using

- Provider name

CUSTOMERUnderstood. You have moved my appointment to Thursday at 04:10PM. Main Campus Building B. And the confirmation code is RVB3M8P2. Yes. That's right. Thank you for your help. Goodbye.

AGENTYou’re very welcome, Arjun. Just to confirm the key details before you go:

- Dermatology

AGENTwith Dr. Okafor - New time: Thursday at 4:10 PM - Location: Main Campus, Building B - Confirmation code: RVB-3M8P2 Please arrive 15 minutes early with a photo ID and your insurance card, and park in free patient Lot C off River Road. Take care, and goodbye.

Home services

AGENTCopperline Home Services, this is Alex. What can I help you with?

CUSTOMERHi. I need help right now. A pipe under my kitchen sink has burst. Water is spreading across the floor. And it's going to water Power Strip. Haven't shut the water off yet. Can you send someone?

AGENTThanks for telling me right away—this can be dangerous. Let’s handle safety first. 1. Stay away from the power strip. Do not touch it. 2. If you can safely reach your main breaker panel without stepping in water, go there now and:

- Switch off the breaker

CUSTOMEROkay. I'll just try to get to the breaker panel. Do you need me to turn it off?

AGENTYes—if you can get there without stepping in any water, please do:

1. Go to your breaker panel. 2. First, switch off the individual breaker for the kitchen, if it’s labeled. 3. If you’re not sure which one that is, or if water is close to outlets/power strip, switch off the main breaker for the whole house. Do not touch the power strip or any cords near the water. Once you’ve flipped the breaker(s), step back to a dry area and

CUSTOMERAlright.

CUSTOMEROkay. I'm doing that now. The power is off. The water is still spreading, though. What should I do next?

AGENTGood—that was the most important step. Next, we want to stop the water:

1. Shut off the water supply- Look under the kitchen sink for two small shutoff valves (one for hot, one for cold) on the pipes going up to the faucet. - Turn each valve clockwise until it stops. - If those valves are missing or won’t turn, do you know where your main water shutoff is (often in a basement, utility room, garage, or outside near the street in a box)?

CUSTOMERUh-huh. I see the valves under the sink. I'm setting them clockwise now. Okay. The water has stopped. Oh, I know where the main shut off is, but I think the sink valve works. Can you send someone to fix the pipe?

AGENTYes, we can absolutely send someone. First, a couple of quick safety/cleanup steps while we get you on the schedule: - Keep everyone away from that wet area and the power

CUSTOMERRight. I'll make sure nobody goes near it. So, can you tell me how much the visit will cost?

AGENTYou’re all set on keeping folks clear of that area—thank you for doing that. I’ve got your account here, and you’re listed as a Copperline Care member at:

- 1480 Larkspur Avenue, Unit 4B

Because this is an emergency same-day visit, your callout fee as a Copperline Care member is $99. Any repair work or parts would be additional and the technician will go over pricing with you on-site before starting. I’ve already opened an emergency ticket and booked the earliest window

- Ticket numberCPL-58210 - Arrival window: today, 2:00 PM to 4:00 PM - Address on file: 1480 Larkspur Avenue,

CUSTOMERI actually moved from Unit 4 B to Unit 4 D. Last month. So it's 1480 Larkspur Avenue, Unit 4 D. Can anyone come before 2PM? This is urgent. And who will be coming, and how will I know they're on the way?

AGENTThanks for updating that—good to know you’re in 4D now. I’ve updated your service request to: - 1480 Larkspur Avenue, Unit 4D On your other questions:

1. Sooner than 2 PMI checked our emergency schedule before: the earliest emergency arrival window available is still today, 2:00 PM–4:00 PM. There isn’t anything sooner showing in the system, even for urgent calls. If something earlier opens up, dispatch will prioritize you because this is already marked as an emergency

CUSTOMERAlright. Got it. So, 2PM to 4PM is the earliest. Will be coming? And how will I know they are on the way?

AGENTYou’re correct: 2:00 PM–4:00 PM today is the earliest emergency window we have you booked for. Here’s what to expect:

- Who will be comingA Copperline plumbing technician assigned by our dispatch team. - How you’ll know they’re on the way: Our system notes: “The technician texts the callback number 30 minutes before arrival.” So you’ll get a text to your phone about half an hour before they show up. Let me read back what I have for you: - Service address: 1480 Larkspur Avenue, Unit 4D - Ticket number: CPL-58210 - Arrival window: today, 2:00 PM–4:00 PM - Emergency callout fee (Copperline Care member): $99, plus any repair/parts, which the tech will quote before starting.

Is there a gate code or special access info they need to reach Unit 4D?

CUSTOMERYes. There's a gate code. Pound 2719. And, the callback number is (503) 555-0139. No. 0193. That's right. I've got the ticket number. CPL58210. For your help. Goodbye.

Insurance

AGENTNorthgate Mutual claims, this is Alex. Are you somewhere safe right now?

CUSTOMERUh-huh. Yes. I am standing next to my car on the sidewalk. Nobody is hurt. I am fine.

AGENTI’m glad to hear you’re okay and that no one is hurt. Let’s get your claim started. I’ll need your auto policy number and the policyholder’s last name to look up your policy. Can you read me the policy number and tell me the last name on the policy?

CUSTOMERYeah. Sure. The policy number is NG4471903 No. Sorry. Last one is two. The last name is Ramirez.

AGENTThank you, Ms. Ramirez. I’ve pulled up your policy ending in 1902 for your 2021 Honda CR‑V. Now I’ll get some details about what happened. 1) When did the accident happen? 2) Where did it happen? 3) Can you tell me, in your own words, what happened? We’ll

CUSTOMERMhmm. So the access happened about twenty minutes ago. I was stopped at a red light at S T 5th Avenue in Alder Street. Another car hit the back of my car. I was not moving. The back bumper is crushed. And my rear wheel rubs now, so I cannot drive it. Police did not come.

AGENTI’ve opened a claim for you and we’ll get your car taken care of. Here’s what I have so far—stop me if anything is wrong:

-

CUSTOMEROkay. Go ahead.

AGENTHere’s what I have documented:

- Policy numberNG4471902 for Ramirez, 2021 Honda CR‑V - Accident time: about 20 minutes ago - Location of accident: SW 5th Avenue and Alder Street - What happened: You were stopped at a red light, not moving, when another car hit the back of your car. Your rear bumper is crushed and the rear wheel is rubbing, so the car cannot be driven. - Injuries: No one is injured. - Police: Police did not come to the scene. I’ve opened claim number NGM-904417. Your assigned adjuster is Priya Shah, who will contact you within one business day. For the tow: - A truck from Eastside Towing

CUSTOMERMhmm. Got it. Will Eastside Towing take my car right away, or do I have to wait? And, where will they take it?

AGENTEastside Towing is already dispatched for you. - They’re coming directly to you at SW 5th and Alder, with an estimated arrival in about 45 minutes. You don’t need to arrange anything

CUSTOMERAlright. Thank you. And, will they take the car to a repair shop or somewhere else?

AGENTThey’ll take it straight to a repair shop.

CUSTOMEROkay.

AGENTThey’re

CUSTOMERMhmm. That's good. So can you tell me how much do I have to pay for all this? Will there be a deductible? And do I get a rental car while my car is in the shop? Also, is my premium gonna go up because of this?

AGENTHere’s what I can see on your policy and what I can’t answer from here:

1) Your deductible Your policy shows a $500 collision deductible. - That’s the amount you’d be responsible for toward the repairs if this goes through your collision coverage. - If the other driver is found at fault and their insurance pays for everything, your deductible may be reimbursed or not charged, but that depends on how liability is decided and how the claim is handled by your adjuster. 2) Tow cost Your policy includes towing coverage up to 25 miles. - As long as the tow is within that limit (which it is, to our partner shop at 9th and Morrison), the tow itself should be covered under your policy. - You shouldn’t have to pay the tow driver out of pocket for that standard tow. 3) Rental car You do have rental reimbursement coverage$40 per day, up to 30 days, starting from the day

CUSTOMERMhmm. That's helpful. But, he hit me so I should not have to pay the deductible. Right? And about the rental car, how do I get that? Do you arrange it or do I call someone? Also, can you tell me if my premium will go up from this accident?

AGENTI’ll walk through each of those, based only on what I’m allowed to see and explain from here.

CUSTOMERAlright. Go ahead.

AGENTOn the deductible, rental, and premium, here’s how it works and what I can and can’t tell you from here:

1) Deductible when you were rear‑ended Based on your policy, your collision deductible is $500. Whether you actually end up paying that depends on how liability is decided- If the other driver is confirmed at fault and their insurance pays for the damage, your deductible may not apply, or it may be

CUSTOMERUh-huh. Got it. So if the other driver is found at fault, I should not have to pay the deductible. Correct? And for the rental car, do you set it up for me, or should I contact the rental company myself? Also, what about my premium? Will it increase because of this claim?

AGENTOn all three of those, here’s what I can and can’t say from here:

1) Deductible if the other driver is at fault I can’t make the final call on fault or payments—that’s your adjuster’s job—but generally- Your policy has a $500 collision deductible. - If the other driver is confirmed at fault and their insurance fully pays for your damage, you typically would not end up paying that deductible. - If we have to use your collision coverage first, you may see the deductible applied up front, and then your adjuster will pursue the other insurer. If they recover the money, your deductible may be reimbursed. Your adjuster, Priya Shah, will explain exactly how this will work once she’s spoken with you and the other company. 2) How to get the rental car Your policy includes rental reimbursement: $40 per day, up to 30 days, from the day the claim is opened. There are usually two ways this is handled: - Direct-bill arrangement: In many cases, the adjuster can set you up with a preferred rental company so the billing goes directly through Northgate up to your daily limit. - Reimbursement: Or you may rent a car yourself and submit receipts, and we reimburse you up to $40 per day. From where I am, I can’t actually book the rental for you, but Priya can: - When she calls you (within one business day), tell her you need a rental right away.

CUSTOMERMhmm. I understand. So I should wait for free Priya to call me, and she will arrange the rental car for me. And about the premium, can you see if it will go up, up, this claim, or is that something Pria will tell me too?

AGENTYou’ve got it right about Priya and the rental. On your premium:

- I can’t see or decide future premium changes from here. - Whether your rate goes up depends on several things- Who is found at fault - The total cost of the claim - Your prior claims/accident history - Northgate’s rating rules at renewal time Those decisions are not made at the time of the accident and not by me. Your adjuster, Priya, can talk more specifically once she has all the facts, but even she can’t promise exactly what your renewal premium will be, because that’s set later by underwriting/rating. So:

-

CUSTOMERUh-huh. I see. So, basically, it depends on who is at fault and how the claim goes. Will ask Priya more when to call. And, can you repeat the claim number for me just to make sure I have it

How these numbers are measured →