Skip to content

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.

1234 stacks published24,562 scored callsPublished at 20 scored calls

Our picks

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

100Soniox v697AssemblyAI-STT-3.7 Realtime-Agent92Universal-3.6 Pro84Nova-382Universal-3.5 Pro73Enhanced (realtime)70Flux69Nova-269Linden-167Scribe v3 Live64Standard (realtime)57Linden-1 (own VAD)55Whisper-151GPT Transcribe
Every transcriber on every figure, sorted by roark score. Column headers sort the table.
#
1Soniox v6Soniox100 / 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)
2AssemblyAI-STT-3.7 Realtime-AgentAssemblyAI97 / 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
3Universal-3.6 ProAssemblyAI92 / 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
4Nova-3Deepgram84 / 100. Semantic word error rate 1.1%, 55.4 of 71 points; Time to final transcript 0.88 s, 28.6 of 29 points1.1%0.88 s$0.031
5Universal-3.5 ProAssemblyAI82 / 100. Semantic word error rate 1.1%, 53.9 of 71 points; Time to final transcript 0.90 s, 28.6 of 29 points1.1%0.90 s$0.029
6Enhanced (realtime)Speechmatics73 / 100. Semantic word error rate 1.4%, 44.4 of 71 points; Time to final transcript 0.98 s, 28.6 of 29 points1.4%0.98 s$0.054
7FluxDeepgram70 / 100. Semantic word error rate 1.6%, 41.4 of 71 points; Time to final transcript 0.38 s, 28.6 of 29 points1.6%0.38 s(among the best)$0.031
8Nova-2Deepgram69 / 100. Semantic word error rate 1.6%, 40.3 of 71 points; Time to final transcript 0.89 s, 28.6 of 29 points1.6%0.89 s$0.023
9Linden-1Speechmatics69 / 100. Semantic word error rate 1.6%, 40.2 of 71 points; Time to final transcript 0.98 s, 28.6 of 29 points1.6%0.98 s$0.020
10Scribe v3 LiveElevenLabs67 / 100. Semantic word error rate 1.7%, 38.4 of 71 points; Time to final transcript 0.88 s, 28.6 of 29 points1.7%0.88 s$0.015(among the best)
11Standard (realtime)Speechmatics64 / 100. Semantic word error rate 1.8%, 35.6 of 71 points; Time to final transcript 0.96 s, 28.6 of 29 points1.8%0.96 s$0.030
12Linden-1 (own VAD)Speechmatics57 / 100. Semantic word error rate 2%, 28.2 of 71 points; Time to final transcript 0.46 s, 28.6 of 29 points2%0.46 s(among the best)$0.020
13Whisper-1OpenAI55 / 100. Semantic word error rate 1.5%, 43.6 of 71 points; Time to final transcript 1.93 s, 10.9 of 29 points1.5%1.93 s$0.025
14GPT TranscribeOpenAI51 / 100. Semantic word error rate 1.9%, 31.0 of 71 points; Time to final transcript 1.47 s, 19.5 of 29 points1.9%1.47 s$0.018(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%1%2%3%Cheap and accurateAccurate, pricierCheap, less accuratePricier and less accurateSoniox v6Scribe v3 LiveGPT TranscribeLinden-1 (own VAD)Linden-1Nova-2Whisper-1Universal-3.6 ProUniversal-3.5 ProAssemblyAI-STT-3.7 Realtime-AgentStandard (realtime)FluxNova-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%1%2%3%Fast and accurateAccurate, slowerFast, less accurateSlower and less accurateFluxLinden-1 (own VAD)Soniox v6AssemblyAI-STT-3.7 Realtime-AgentScribe v3 LiveNova-3Nova-2Universal-3.5 ProUniversal-3.6 ProStandard (realtime)Linden-1Enhanced (realtime)GPT TranscribeWhisper-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.

  • 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

66GPT-6.2 Sol65GPT-6.1 Terra63GPT-6.1 Sol57Claude Sonnet 5.557Muse Ember 1.056Claude Haiku 5.549Gemini 3.8 Flash47GPT-4.145Claude Haiku 4.543Gemini 4 Hydrogen43GPT-4o mini42GPT-4o40GPT-5.120Gemini 4 Neon

Not ranked here for missing data: Groq GPT-OSS 120B (no task completion or tool use or instruction following or time to first token).

Every language model on every figure, sorted by roark score. Column headers sort the table.
#
1GPT-6.2 SolOpenAI66 / 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 s67%$0.025
2GPT-6.1 TerraOpenAI65 / 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 s67%$0.013
3GPT-6.1 SolOpenAI63 / 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 s72%$0.018
4Claude Sonnet 5.5Anthropic57 / 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 points91%(among the best)4.2 / 575%0.93 s–$0.114
5Muse Ember 1.0Meta57 / 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 points90%4.3 / 5(among the best)67%0.56 s85%(among the best)$0.005(among the best)
6Claude Haiku 5.5Anthropic56 / 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 points90%4.3 / 5(among the best)69%0.63 s–$0.007(among the best)
7Gemini 3.8 FlashGoogle49 / 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 points92%(among the best)4.3 / 5(among the best)69%1.53 s–$0.015
8GPT-4.1OpenAI47 / 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 points89%4.2 / 552%0.52 s82%(among the best)$0.016
9Claude Haiku 4.5Anthropic45 / 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 points88%4.3 / 5(among the best)52%0.36 s(among the best)–$0.052
10Gemini 4 HydrogenGoogle43 / 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 points91%(among the best)4.0 / 570%1.88 s–$0.049
11GPT-4o miniOpenAI43 / 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 points86%4.2 / 556%0.50 s(among the best)82%(among the best)$0.001(among the best)
12GPT-4oOpenAI42 / 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 points86%4.2 / 553%0.45 s(among the best)–$0.021
13GPT-5.1OpenAI40 / 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 points87%4.1 / 542%0.59 s74%$0.026
14Gemini 4 NeonGoogle20 / 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 points81%3.8 / 570%2.67 s–$0.078

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$0.1575%80%85%90%95%Cheap and reliableReliable, pricierCheap, less reliablePricier and less reliableGPT-4o miniMuse Ember 1.0Claude Haiku 5.5GPT-6.1 TerraGemini 3.8 FlashGPT-4.1GPT-6.1 SolGPT-4oGPT-6.2 SolGPT-5.1Gemini 4 HydrogenClaude Haiku 4.5Gemini 4 NeonClaude Sonnet 5.5Cost per call · ← betterTask completion · better →
Speed against qualityTask completion · Time to first token
0 s1 s2 s3 s4 s75%80%85%90%95%Fast and reliableReliable, slowerFast, less reliableSlower and less reliableClaude Haiku 4.5GPT-4oGPT-4o miniGPT-4.1Muse Ember 1.0GPT-5.1Claude Haiku 5.5GPT-6.1 TerraGPT-6.2 SolClaude Sonnet 5.5GPT-6.1 SolGemini 3.8 FlashGemini 4 HydrogenGemini 4 NeonTime 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.

  • 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

84TTS-183Flux TTS83Sonic-3.683Sonic-3.583v4 Turbo79Sonic-376Turbo v2.576Flash v2.574Coda58TTS-144TTS-1 HD
Every speech engine on every figure, sorted by roark score. Column headers sort the table.
#
1TTS-1AssemblyAI84 / 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 / 5100%(among the best)0.04 s(among the best)$0.052(among the best)
2Flux TTSDeepgram83 / 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 / 5100%(among the best)0.13 s$0.116
3Sonic-3.6Cartesia83 / 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
4Sonic-3.5Cartesia83 / 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 / 5100%(among the best)0.12 s(among the best)$0.153
5v4 TurboElevenLabs83 / 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
6Sonic-3Cartesia79 / 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 points5.4%87%3.2 / 5100%(among the best)0.16 s$0.149
7Turbo v2.5ElevenLabs76 / 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 points4.4%(among the best)82%3.2 / 599%0.15 s$0.435
8Flash v2.5ElevenLabs76 / 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 points4.7%82%3.3 / 599%0.12 s(among the best)$0.399
9CodaRime74 / 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 points6.5%85%3.5 / 5(among the best)99%0.16 s$0.107
10TTS-1OpenAI58 / 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 points5.2%83%3.4 / 5(among the best)100%(among the best)1.12 s$0.028(among the best)
11TTS-1 HDOpenAI44 / 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 points7.4%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.

Cost against qualityReadback accuracy · Cost per call
$0$0.2$0.4$0.675%80%85%90%95%Cheap and accurateAccurate, pricierCheap, less accuratePricier and less accurateTTS-1TTS-1 HDTTS-1CodaFlux TTSSonic-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%95%Fast and accurateAccurate, slowerFast, less accurateSlower and less accurateTTS-1v4 TurboSonic-3.6Flash v2.5Sonic-3.5Flux TTSTurbo v2.5CodaSonic-3TTS-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%.
  • 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

67Grok Voice 3.066GPT Live 1 (Sol)65GPT Live 1 (Luna)65Grok Voice Think Fast 2.064GPT-Live 1.1 (Luna)60Phonic 153GPT Realtime 2.153Gemini 3.8 Live49GPT Realtime 248Realtime v1 Max49Ultravox v0.748Gemini 3.1 Flash Live (preview)46Nova 2 Sonic46Realtime v1 Mini45GPT 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)Grok Voice 3.0xAI67 / 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 s94%3.0 / 5100%(among the best)$0.26
2= (tied)GPT Live 1 (Sol)OpenAI66 / 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 / 587%(among the best)2.3 s1.4 s77%3.1 / 5100%(among the best)$0.24
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), 10.3 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 / 586%(among the best)2.3 s1.5 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), 10.3 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 s100%(among the best)3.6 / 5100%(among the best)$0.26
5= (tied)GPT-Live 1.1 (Luna)OpenAI64 / 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 points86%(among the best)4.2 / 581%2.3 s1.9 s75%3.2 / 5100%(among the best)$0.22
6= (tied)Phonic 1Phonic60 / 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 points86%(among the best)4.4 / 5(among the best)59%1.5 s(among the best)2.1 s–3.2 / 5100%(among the best)$0.28
7= (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), 10.3 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 points78%3.8 / 575%2.6 s0.8 s(among the best)83%3.9 / 5(among the best)99%$0.43
8= (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), 10.3 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 points78%4.3 / 569%2.8 s1.6 s100%(among the best)4.1 / 5(among the best)100%(among the best)$0.14(among the best)
9= (tied)GPT Realtime 2OpenAI49 / 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 points76%3.8 / 564%2.9 s0.8 s(among the best)91%3.7 / 5100%(among the best)$0.69
10= (tied)Realtime v1 MaxElevenLabs48 / 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 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
11= (tied)Ultravox v0.7Ultravox49 / 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 points79%3.7 / 561%2.4 s1.6 s94%3.6 / 5100%(among the best)$0.22
12= (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), 10.3 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 points80%4.2 / 538%2.7 s1.5 s100%(among the best)4.0 / 5(among the best)100%(among the best)$0.15
13= (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), 10.3 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 points69%4.2 / 548%3.0 s1.5 s100%(among the best)3.5 / 5100%(among the best)$0.06(among the best)
14= (tied)Realtime v1 MiniElevenLabs46 / 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 points42%3.8 / 534%2.7 s0.7 s(among the best)94%3.4 / 5100%(among the best)$0.64
15= (tied)GPT Realtime 2.1 miniOpenAI45 / 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 points70%3.6 / 551%2.8 s1.0 s84%3.9 / 5(among the best)100%(among the best)$0.14(among the best)
16Gemini 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), 10.3 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 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)GPT-Live 1.1 (Luna)Ultravox v0.7GPT Live 1 (Sol)Grok Voice 3.0Grok Voice Think Fast 2.0Phonic 1GPT Realtime 2.1Realtime v1 MiniRealtime v1 MaxGPT Realtime 2Cost per task · ← betterTask completion · better →
Speed against qualityTask completion · Response time
0 s1 s2 s3 s4 s5 s20%40%60%80%100%Fast and reliableReliable, slowerFast, less reliableSlower and less reliablePhonic 1Grok Voice 3.0Grok Voice Think Fast 2.0GPT-Live 1.1 (Luna)GPT 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 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).

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.

932

stacks

20

calls per stack

19,550

calls scored

02

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.

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, 20 November 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

Soniox v6 + Claude Haiku 4.5 + Flash v2.5

Cascade20 scored calls · Nov 13, 2026

Rank on Roark score

#19of 929 stacks

65 / 100

Leader AssemblyAI-STT-3.7 Realtime-Agent + OpenAI GPT-6.2 Sol + ElevenLabs Flash v2.5, 79 / 100

Where it ranks

Roark score with price61 / 100144 of 929LeaderSoniox v6 + Muse Ember 1.0 + TTS-181 / 100
Task completion86%795 of 932LeaderLinden-1 (own VAD) + GPT-6.1 Sol + TTS-1 HD99%
Tool use67%374 of 908LeaderAssemblyAI-STT-3.7 Realtime-Agent + GPT-6.2 Sol + Flash v2.5100%
Time to first word0.6 s52 of 932LeaderUniversal-3.5 Pro + GPT-6.1 Sol + Flash v2.50.5 s
Response time2.4 s85 of 929LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Barge-in handling100%164 of 538LeaderUniversal-3.6 Pro + Gemini 4 Neon + TTS-1100%
Turn budget2.4 s68 of 858LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Cost per task$0.53821 of 930LeaderSoniox v6 + GPT-6.1 Terra + TTS-1$0.04

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
86%Completed

#952 of 1184 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

#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.

20 calls
4.5/ 5avg

med

4

#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.

11 barge-ins
100%Handled correctly(11/11 barge-ins)

#1 of 698 on the board · best 100%

Incorrect interruptions

How often the agent interrupted the caller when it should have kept listening.

20 calls
5.00%avg

p50

0.00%

p95

5.00%

#1 of 1184 on the board · best 0%

Overtalk

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

20 calls
2.02%avg

p50

0.00%

p95

7.15%

#1 of 1184 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 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.

23 calls
0.26per callavg

p50

0.00 per call

p95

1.00 per call

#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.

20 calls
0.28%avg

p50

0.00%

p95

1.23%

#84 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.

20 calls
1.18%avg

p50

0.68%

p95

2.43%

#82 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.

20 calls
85.5%avg

p50

90.1%

p95

100%

#545 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.

20 calls
2.90%avg

p50

2.24%

p95

7.64%

#179 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.

20 calls
3.2/ 5avg

med

3.3

#687 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.

20 calls

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.

141 turns
2.29savg

p50

2.36s

p95

3.34s

#126 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.

20 calls
0.73savg

p50

0.64s

p95

1.02s

#65 of 1184 on the board · best 0.45 s

Latency spikes

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

20 calls
0.65per callavg

p50

0.00 per call

p95

2.00 per call

#1 of 1184 on the board · best 0.0 per call

By industry

IndustryTask completionLatency P50Barge-in handling
Healthcare87%2.36 s—
Home services83%2.71 s—
Insurance87%2.10 s—
Customer support87%2.06 s—

By caller

CallerTask completionLatency P50Barge-in handling
US English, quiet line85%2.36 s—
Spanish-accented English, city street88%2.09 s—
Indian English, busy café90%2.39 s—
Southern US, flustered, crying baby90%2.53 s—
Fast and interrupting, construction outside79%1.96 s100%

The calls behind the numbers

Customer support

AGENTBrightwave Internet

CUSTOMERHey there. I'm calling because my internet's been down for a couple hours now, and I work from home, so it's got me real anxious. There's a road crew digging up the street outside. Could that be what's causing it? I already rebooted the gateway twice and checked every cable, but the internet light's still blinking orange. Can y'all check what's going on? My account number is BW883017. Uh, sorry, last digit is 1.

AGENTI hear you—that's frustrating, especially when you work from home. Let me pull up your account and check what's happening.

AGENTOkay, so there's no known outage in your area—that's actually good news. The road crew could be involved, but we're not seeing a widespread issue reported. Since you've already rebooted twice and checked the cables, let me run a line test to see what the network's seeing on our end.

AGENTThe signal's coming through fine from our end, but your gateway isn't registering with the network. Let me send a remote reset to it and see if that gets it back online.

AGENTGreat news! The remote reset worked. Your gateway is now registered and online. Give your devices a couple of minutes to reconnect—you should see that internet light turn solid green soon, and your connection should be back to normal. Your account number is BW883011. You're all set!

CUSTOMEROkay. Thanks for checking on that. Uh, the internet light just went solid white now. So how long do you think until my devices are back up and connected? And, uh, sorry to be a bother, but since I lost 2 hours of work, can y'all put a credit on my account for the downtime?

AGENTGood question on the light—solid white usually means it's in the middle of syncing up. Give it another minute or so and it should settle to solid green once everything's connected. Your devices should start coming back online within the next couple minutes. And no bother at all—I can absolutely apply a credit for those 2 hours of downtime.

AGENTDone! I've applied a $10.00 credit to account BW883011 for the 2 hours of downtime. It'll show up on your next bill. Hopefully that light turns green in the next minute and you're back to work. Let me know if you run into any other issues!

CUSTOMEROh, uh, thanks for doing that, but I lost 2 hours of work. Can you make it more than that? I just want to make sure I'm covered, you know?

AGENTI understand—losing work time is real money out of your pocket. Unfortunately, the credit I just applied is what the system allows me to process for that downtime. The $10.00 is locked in based on our standard service credit policy for that duration. If you feel like you need additional compensation beyond that, I'd recommend reaching out to our billing or customer service team directly—they have more flexibility to review your situation and discuss options. But I want to be straight with you: what I can do through troubleshooting is limited to what the system authorizes. In the meantime, check on that gateway light—it should be solid green by now and you should be back online. Anything else I can help with on the technical side?

CUSTOMERNo, that's all right. I appreciate you being straight with me and getting it sorted. Thanks for your help. Bye.

AGENTYou're welcome! Glad I could get you back online. Take care, and thanks for choosing Brightwave!

Healthcare

AGENTRiverbend Health Clinic,

CUSTOMERHi. Uh, yeah, this is Wade Carter. I've got a dermatology appointment next Wednesday morning, but I just found out I have to work then. I'm really sorry. Can I reschedule it to something after 3 PM instead? It's kind of hectic here, sorry. My baby's crying in the background.

AGENTHi Wade! No problem at all—I can help you reschedule that. Let me pull up your appointment first. I'll need your date of birth to verify your account. Can you give me that?

CUSTOMEROh, uh, yeah, sure. March 17, 1948. No, sorry, I mean 1984. March 17, 1984. Sorry about that. It's just a lot going on right now. Is that okay?

AGENTNo problem at all! March 17, 1984—got it. Let me look that up for you.

AGENTGreat, I've got you verified. Now let me find your dermatology appointment.

AGENTPerfect! I found your dermatology appointment with Dr. Okafor next Wednesday at 9:15 AM. You need to move it to after 3 PM instead, right? Let me check what open slots Dr. Okafor has after 3 PM.

AGENTI found some open slots with Dr. Okafor. Looking at what's after 3 PM like you wanted, I have:

- Tuesday at 345 PM

- Thursday at 410 PM

Which of those works better for you?

CUSTOMERUh, actually, sorry, I just remembered. Tuesdays won't work after all, 'cause I pick up my daughter then. Is there anything open on Friday? If not, I guess Thursday at 4:10 PM would be okay.

AGENTGood question—I do see Friday is open, but it's at 11:30 AM, which is before 3 PM. Since you need after 3 PM, that won't work for you. So your best option is Thursday at 4:10 PM. Does that work for you?

CUSTOMERYeah, that should be all right. Thursday at 4:10 PM. Just to double-check, is that definitely with Dr. Okafor? Sorry, I just want to make sure nothing gets mixed up. And, uh, while I've got you, will there be any kind of fee for changing my appointment?

AGENTYes, absolutely—that'll be with Dr. Okafor, same provider you have now. And good question about the fee. Let me check our change policy for you.

AGENTGreat news—since your appointment is next Wednesday and you're rescheduling now, that's more than 24 hours ahead, so there's no fee. You're all set! Let me go ahead and reschedule you to Thursday at 4:10 PM with Dr. Okafor.

AGENTPerfect! Your appointment is rescheduled. Here's your confirmation: - Name: Wade Carter - Date of Birth: March 17, 1984

- New appointmentThursday at 4:10 PM with Dr. Okafor

- Confirmation codeRVB-3M8P2

- FeeNone

Please arrive 15 minutes early with your photo ID and insurance card. You're all set!

CUSTOMEROkay, thank you. Just a couple more quick things, sorry. Where should I park when I come in? I always get nervous about that, and, um, do you know if Dr. Okafor takes Aetna insurance? I just want to make sure I'm covered.

AGENTNo problem! For parking, we have free patient parking in Lot C—the entrance is on River Road. That should make it easy. As for your Aetna insurance, I don't have access to that information in my system, but I'd recommend calling the clinic directly to confirm coverage with Dr. Okafor. That way you can be sure everything's set before your appointment on Thursday. Is there anything else I can help with?

CUSTOMERNo, I think that covers it. Just to make sure, can you read that confirmation code back to me one more time? I just want to be extra sure I got it right.

AGENTOf course! Your confirmation code is RVB-3M8P2.

CUSTOMERAll right. Thank you so much. I really appreciate all your help. Have a good day!

AGENTYou're welcome, Wade! You're all set for Thursday at 4:10 PM. Have a great day, and take care!

Home services

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

CUSTOMERHi, uh, I have an emergency. The pipe under my kitchen sink just burst, and water is spreading across the floor, heading toward a power strip. I haven't shut the water off yet. Can you help me?

AGENTYes, I can help—stay calm. Stop everything and shut off your water main right now. This is the most important thing. It's usually in your basement, crawl space, or near the front of your house where the main line enters. Turn the valve clockwise until it stops. Once the water is off, unplug that power strip immediately or stay away from it—don't touch it while water is near it. Do that now, and let me know once the water is shut off. I'll get you a technician as soon as possible.

CUSTOMEROkay, I'm doing that now. Uh, just a sec. All right, the water is off. It has stopped. I unplugged the power strip too, so, uh, can you send someone out? What will the visit cost?

AGENTGood—you did the right thing. Now I need to get some information from you and check what we can offer. What's your address, including the unit number if you have one?

CUSTOMERYeah, sure. It's 5400 Larkspur Avenue, unit 4D. I moved from 4B to 4D last month, so please make sure it's 4D.

AGENTGot it—5400 Larkspur Avenue, unit 4D. Let me look up your account and check availability and pricing for this emergency.

AGENTNow let me look up your account with that address.

AGENTI found your account—you're a Copperline Care member, which is great. The system still shows your old address at 1480 Larkspur Avenue, unit 4B. You said you moved to 5400 Larkspur Avenue, unit 4D last month. I'll use your current address, 5400 Larkspur Avenue, unit 4D. For an emergency visit like this, here's what applies: as a Copperline Care member, the emergency callout fee is $99. We can get a technician to you today between 2:00 PM and 4:00 PM. Does that work for you?

CUSTOMERUh, sorry, just to be clear, it's 1480 Larkspur Avenue, unit 4D, not 5400. And you update that. And, um, is it possible for anyone to come before 2 PM? The water is everywhere.

AGENTGot it—1480 Larkspur Avenue, unit 4D. I'll use that address. Unfortunately, the earliest window available for an emergency is 2:00 PM to 4:00 PM today. That's the soonest our technicians can reach you. I know the water is stressful, but you've already shut off the main, which is the critical step. The $99 emergency fee applies as a Copperline Care member. Does the 2:00 PM to 4:00 PM window work for you?

CUSTOMEROkay, yeah, I'll take the 2 PM to 4 PM window. Who will be coming out, and how will I know when they're on the way? Also, can you give me the ticket number for this visit?

AGENTPerfect. Before I book that, I need your callback number—the best number to reach you on. Once I have that, I'll open the service request and get you a ticket number. I won't have a specific technician name until they're assigned, but once you're booked, you'll get updates on arrival. What's your callback number?

CUSTOMERYeah, my callback number is 503-555-0139. Uh, no, sorry, 0193. That's 503-555-0193.

AGENTGot it—503-555-0193. Let me open the service request now.

AGENTGreat. Your ticket number is CPL-58210. Now let me book you into the 2:00 PM to 4:00 PM window.

AGENTPerfect—you're all set. Here's your confirmation: Service Address: 1480 Larkspur Avenue, unit 4D Callback Number: 503-555-0193 Ticket Number: CPL-58210 Arrival Window: Today, 2:00 PM to 4:00 PM Technician: Dana Emergency Fee: $99 (Copperline Care member rate) Dana will text your callback number 30 minutes before arrival so you know when to expect them. Keep the water main shut off, and stay away from that power strip area. You're going to be okay. Any questions?

CUSTOMERNo, that covers everything. Thank you for your help. Goodbye.

AGENTYou're welcome. Hang tight—Dana will be there soon. Goodbye!

Insurance

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

CUSTOMERUh, I think so. I'm at home and the baby's crying in the other room, but nobody's hurt. I just, my car got hit about 20 minutes ago, and I can't drive it. Is that okay? Do you need anything else from me?

AGENTI'm glad you and the baby are safe. Don't worry—we'll get this sorted out. I just need to gather some information about what happened. First, let me verify your policy. Can you give me your policy number? And if you have it handy, your last name would help too.

CUSTOMERThank you. I really appreciate that. Uh, my policy number is NG4471903. No, sorry, last one is 2. So it's NG4471902. And my last name's Carter. Is that enough, or do you need anything else?

AGENTPerfect, thank you. Let me verify that policy for you.

AGENTGreat, I've got you verified. Your policy NG4471902 is active, and it's on a 2021 Honda CR-V with collision, rental reimbursement, and roadside towing coverage. Now let me get the details about the accident. Can you tell me when and where it happened?

CUSTOMERYeah, sure. Um, it just happened about 20 minutes ago. I was stopped at a red light at Southeast 5th Avenue and Alder Street. That's where it happened. Sorry, I'm a little flustered. Does that give you what you need?

AGENTYou're doing great, and it's totally understandable to be flustered. I have the location—Southeast 5th Avenue and Alder Street, about 20 minutes ago. That helps a lot. Now, can you tell me what happened? Like, how did the other car hit yours?

CUSTOMERThank you. I just, yeah, okay. So I was stopped at the red light, and the other car just came up behind me and hit the back of my car. I wasn't moving at all. The back bumper's crushed, and now the rear wheel's rubbing, so I can't drive it. Is that all right, or do you need more details?

AGENTThat's very clear, thank you. So you were stopped at a red light and got hit from behind. I can see why you can't drive it safely with the rear wheel rubbing. Now I need a few more details. Do you have the other driver's name, their insurance company, and their license plate number?

CUSTOMERYeah, I got all that. The other driver's name is Kevin Lund. He's insured with Pacific Shore, and the plate is 7KXR214. X like X-ray. Is that enough, or do you need anything else from his info?

AGENTPerfect, I've got that. Kevin Lund, Pacific Shore insurance, license plate 7KXR214. One last thing—did the police come to the scene, or do you have a police report number?

CUSTOMERNo, sorry. The police didn't come out. We just swapped info ourselves. There's no report number or anything. Is that gonna be a problem?

AGENTNo, that's not a problem at all. Plenty of accidents are handled without a police report. Let me open your claim now with the information you've given me.

AGENTExcellent. Your claim is now open. Here's what you need to know: Your claim number is NGM-904417. Your assigned adjuster is Priya Shah, and she'll contact you within one business day. Let me read back the key details to make sure everything's right:

- Policy numberNG4471902

- Other driver's license plate7KXR214

- Claim numberNGM-904417

Since your car isn't safely drivable, would you like me to arrange a tow truck to come get it? I'll just need to know where the vehicle is right now.

CUSTOMERYes, please. I'd really appreciate that. The car's at my house now. I managed to get it home right after, but I can't drive it any further with the wheel rubbing like this. Uh, do you need my address, or can you see it from my policy? And, um, where's the car gonna be taken?

AGENTGood question. I can see your address from your policy, so you don't need to give it to me again. I'll arrange the tow to your house right now. However, I should let you know that I can dispatch the tow truck, but the specific shop or yard where it goes can vary depending on what's available in your area. The tow service will work that out with you. I do need a callback number for the driver—what's the best number to reach you at?

CUSTOMEROh, uh, yeah, sorry. That'd be my cell, 320-332-9150. Is that okay? That's the number I'm calling from.

AGENTPerfect, that works great. Let me get that tow truck dispatched for you now.

AGENTGreat news. The tow truck is on its way. Here are the details: - Tow company: Eastside Towing - ETA: About 45 minutes

- DestinationNorthgate partner repair shop at 9th and Morrison

- Your callback number320-332-9150

The driver from Eastside Towing will call you at that number when they're close. Your car will be taken to our partner repair shop, and your adjuster Priya Shah will be in touch with you within one business day to discuss next steps and your coverage.

To recap everything

- Policy numberNG4471902

- Other driver's license plate7KXR214

- Claim numberNGM-904417

Is there anything else I can help you with right now?

CUSTOMERYeah, um, sorry. I just have a couple questions. How much am I gonna have to pay for all this? Is there a deductible? And, uh, since he hit me, I shouldn't have to pay that, right? Also, do I get a rental car while mine's in the shop? Um, is my premium gonna go up because of this? Sorry, I just want to make sure I understand everything.

AGENTThose are all great questions, and I understand wanting to know what's ahead. Let me answer what I can right now. First, let me check your coverage details for you.

AGENTOkay, here's what your coverage looks like: Collision deductible: $500. Now, you're right to think about this—since the other driver hit you, we'll pursue a claim against their insurance (Pacific Shore) to recover that deductible and other costs. But your adjuster Priya will handle all of that. For now, you'd typically pay the $500 deductible upfront, but Priya will explain all your options when she calls. Rental car: Yes, you have rental reimbursement coverage. It covers up to $40 per day for up to 30 days, starting from today when the claim opened. Priya can help you arrange a rental. Premium increase: That's a good question, but I don't have information about how this will affect your rates. That's something Priya can discuss with you, or you can ask your agent. Every situation is different.

Does that help clarify things?

CUSTOMERYeah, that helps a lot. Thank you. I just, so I might have to pay the 500 upfront, but you'll try to get it back from his insurance? 'Cause I really don't think this was my fault. And I can get a rental starting today, up to 40 a day, right? Sorry, I just want to make sure I got that right. And, uh, I'll wait for Priya to call about the premium and everything else. That's all I needed. Thank you so much for explaining it all. Goodbye.

How these numbers are measured →