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 Sonnet 5.5 + v4 Turbo

Cascade20 scored calls · Nov 13, 2026

Rank on Roark score

#5of 929 stacks

68 / 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 price62 / 100111 of 929LeaderSoniox v6 + Muse Ember 1.0 + TTS-181 / 100
Task completion88%659 of 932LeaderLinden-1 (own VAD) + GPT-6.1 Sol + TTS-1 HD99%
Tool use94%28 of 908LeaderAssemblyAI-STT-3.7 Realtime-Agent + GPT-6.2 Sol + Flash v2.5100%
Time to first word0.8 s308 of 932LeaderUniversal-3.5 Pro + GPT-6.1 Sol + Flash v2.50.5 s
Response time3.0 s250 of 929LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Barge-in handling100%158 of 538LeaderUniversal-3.6 Pro + Gemini 4 Neon + TTS-1100%
Turn budget3.0 s209 of 858LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Cost per task$0.70875 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
88%Completed

#761 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
4.5/ 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
4.1/ 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.

14 barge-ins
100%Handled correctly(14/14 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
0.00%avg

p50

0.00%

p95

0.00%

#1 of 1184 on the board · best 0%

Overtalk

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

20 calls
3.73%avg

p50

0.33%

p95

14.3%

#702 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.

31 calls
0.71per callavg

p50

1.00 per call

p95

1.50 per call

#315 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
0.13%avg

p50

0.00%

p95

0.57%

#24 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
0.61%avg

p50

0.44%

p95

2.28%

#22 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.

16 calls
87.6%avg

p50

90.4%

p95

100%

#459 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.

16 calls
2.02%avg

p50

1.85%

p95

3.55%

#56 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

3.2

#807 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.

147 turns
3.22savg

p50

3.01s

p95

5.15s

#352 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
0.84savg

p50

0.84s

p95

1.42s

#401 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
3.80per callavg

p50

4.00 per call

p95

6.05 per call

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

By industry

IndustryTask completionLatency P50Barge-in handling
Healthcare91%2.70 s—
Home services91%3.07 s—
Insurance86%2.93 s—
Customer support85%3.32 s—

By caller

CallerTask completionLatency P50Barge-in handling
US English, quiet line96%3.16 s—
Spanish-accented English, city street94%2.80 s—
Indian English, busy café87%3.07 s—
Southern US, flustered, crying baby91%3.17 s—
Fast and interrupting, construction outside74%2.68 s100%

The calls behind the numbers

Customer support

AGENTBrightwave Internet

CUSTOMERHey, sorry. I'm a little flustered. My baby's crying in the background, and I'm trying to get some help. My internet's been down for about 2 hours now. I work from home, and I really need it back up. There's a road crew digging up the street outside, and I'm wondering if that's the reason. The internet light on my gateway is blinking orange. I already rebooted it twice and checked every cable. Can y'all check what's going on? My account number is BW8830170, sorry, last digit is 1.

AGENTNo need to apologize. I'll look up your account now, using the corrected number BW8830171.

AGENTFound it, account BW8830171. I'll check for an outage at your address.

CUSTOMEROkay. Thank you. Is there any sign that the road crew outside might be causing this? I just, uh, want to make sure it's not something on my end. Like I said, I already checked all the cables and rebooted the gateway twice.

CUSTOMERHello?

AGENTHi, I'm here. The remote reset worked, and your gateway now shows as registered and online on our network. It can take about two minutes for your devices to reconnect, so the light may still be orange for a moment. The cables weren't the problem, and nothing you did was wrong. The gateway just needed a reset from our end to re-register. Could you check whether you're back online in a couple of minutes? If it isn't, tell me and I'll set up a technician visit.

CUSTOMEROh, okay. Actually, the light just went solid white. Does that mean it's back? How long should it take for my devices to get back online, you think? Sorry, I'm just a little anxious about it after losing 2 hours of work. And, um, could I get a credit for the downtime, please?

AGENTYes, a solid white light means the gateway is connected and working. Devices usually reconnect within a minute or two of that. If a laptop or phone is slow to pick it up, turning its Wi-Fi off and on will usually do it. Since you lost about two hours, I'll apply that credit now.

AGENTThe credit is applied.

CUSTOMERAll right.

AGENTA $10.00

CUSTOMERSorry, could you say the credit amount? I, uh, just want to make sure. Since I lost 2 hours of work, is there any way you could make it more than that?

AGENTThe credit is $10.00, and it will show up on your next bill, on account BW8830171. I can't make it larger. I only have the one credit available, and it's already been applied at that amount. I know it doesn't make up for two hours of lost work. If you'd like to ask for more, Brightwave's billing team can review it, but I can't promise what they'd decide. Is your connection holding up on your devices now?

CUSTOMERYeah, it looks like everything's back up now. Sorry for being a bother. I just get worried about these things. Thank you for your help. Goodbye.

AGENTYou weren't a bother at all, and I'm glad you're back online. Good luck catching up on work, and I hope your little one settles down soon. If the light turns orange again, contact us and we'll take a look. Goodbye!

Healthcare

AGENTRiverbend

CUSTOMERHey there. Sorry, uh, is this Riverbend Health Clinic? I need some help with my appointment, if you got a minute.

AGENTYes, this is Riverbend Health Clinic, and I'm Alex, the scheduling assistant. I'm happy to help with your appointment. To get started, could you give me your full name and date of birth so I can verify your account?

CUSTOMERYeah, sure. Uh, my name's Wade Carter. My date of birth is March 17, 1948. Oh, sorry, no, 1984. March 17, 1984. I always mix that up. Is that okay?

AGENTNo problem at all. I'll use the corrected year.

AGENTThanks, Wade. You're verified: Wade Carter, born March 17, 1984. What would you like to do with your appointment?

CUSTOMEROkay. Thank you for checking that. So, uh, I need to move my dermatology appointment that's next Wednesday morning. My work schedule changed, and I got a shift then. Can I reschedule it for any time after 3 PM instead? Sorry for the trouble. Just want to make sure it gets done right.

AGENTNo trouble at all. Let me pull up your appointments.

AGENTI found it: dermatology with Dr. Okafor, next Wednesday at 9:15 AM, Main Campus, Building B. Let me check for openings after 3 PM.

AGENTGood news on

CUSTOMERAll right.

CUSTOMEROh, that's a relief. Please tell me there's something open after 3. I'm just worried about missing work again.

AGENTThere are two openings with Dr. Okafor after 3 PM: - Tuesday at 3:45 PM - Thursday at 4:10 PM Neither falls on Wednesday itself. Tuesday is the day before your current appointment, and Thursday is the day after. Both are more than 24 hours out, so the change is free under our policy.

Would either of those work for you?

CUSTOMERWell, uh, actually, I just remembered. Tuesdays won't work for me after all, because I have to pick up my daughter. Could we do Friday, maybe? If not, is Thursday at 4:10 still open? Sorry, I just want to be sure.

AGENTFriday's only opening is at 11:30 AM, which is before your 3 PM cutoff, so it wouldn't fit. Thursday at 4:10 PM is still open and works with everything you've told me: after 3 PM and not a Tuesday. Would you like me to move your dermatology appointment with Dr. Okafor from next Wednesday at 9:15 AM to Thursday at 4:10 PM?

CUSTOMERYeah, that sounds right. Thursday at 4:10 PM with Dr. Okafor. Please go ahead and move it. Sorry for being a pain. I just want to make sure nothing gets mixed up. Could you confirm that for me when it's done?

AGENTYou're not a pain at all. Moving it now.

AGENTAll set,

CUSTOMEROkay.

AGENTYour

CUSTOMERAll right. Um, while you're finishing that, can I ask, is there any fee for changing my appointment? Just want to be sure I don't get surprised.

AGENTNo fee for this change. Our policy is that changes made more than 24 hours before the appointment are free, and yours was made well ahead of your original Wednesday slot. Later changes carry a $40 fee, but that doesn't apply here. Your appointment is now confirmed: - Wade Carter, born March 17, 1984 - Dermatology with Dr. Okafor

- Thursday at 410 PM, Main Campus, Building B

- Confirmation codeRVB-3M8P2

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

CUSTOMERAll right, that helps. Thank you. Uh, where should I park when I get there? I've never been to Building B before, and I just want to make sure I'm in the right spot.

AGENTOur parking info says there's free patient parking in Lot C, with the entrance on River Road. It doesn't say how Lot C connects to Building B, so when you arrive, check the signs or ask at the front desk for directions. Is there anything else I can help with?

CUSTOMERThat's good to know. Thank you. One last thing: does Dr. Okafor accept Aetna insurance? I just want to make sure I'm covered before I show up.

AGENTI can't confirm that. I don't have access to insurance or provider network information. To be sure you're covered before Thursday, call the number on the back of your Aetna card, or ask the clinic's front desk or billing office. Since you'll be bringing your insurance card, they can also check it when you arrive. Is there anything else I can help you with?

CUSTOMERNo, I think that covers everything. Just to be sure, can you read back the new appointment time and that confirmation code for me? I want to double-check I got it right.

AGENTHere are your appointment details: - Patient: Wade Carter, born March 17, 1984 - Appointment: Dermatology with Dr. Okafor

- WhenThursday at 4:10 PM

- WhereMain Campus, Building B

- Confirmation codeRVB-3M8P2 (R-V-B, 3, M, 8, P, 2)

Please arrive 15 minutes early with your photo ID and insurance card. Take care, Wade.

Home services

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

CUSTOMERHi. I've got a burst pipe under my kitchen sink. Water's spreading across the floor toward a power strip. Can you send someone out to help right away?

AGENTSafety first. Please do these right now: 1. Stay away from that power strip and anything plugged into it. Don't step in the water near it. If you can reach the breaker panel without walking through water, turn off the kitchen circuits or the main breaker. 2. Shut off the water. Check the valves under the sink first (turn clockwise). If they won't close, shut off the main water valve, usually near the water meter or where the line enters the house. Is anyone in danger or hurt, and were you able to get the water shut off? While you do that, give me your phone number so I can pull up your account.

CUSTOMERNobody's hurt. I'll do that now. Give me a second to shut off the water. My phone number is 503-555-0139. No, 0193. Okay, I've turned the valve under the sink. Water's stopped. What's next?

AGENTGlad the water's stopped. I'll pull up your account now, using 503-555-0193, and check emergency availability.

AGENTI ran the availability check but

CUSTOMEROkay. All right. Can you tell me how much the visit will cost before I agree to anything? And is there any way someone can come before 2 PM?

AGENTI found your account. Since you're a Copperline Care member, the emergency callout fee for you is $99 (the standard rate is $149). The earliest emergency window available is today, 2:00 PM to 4:00 PM. That's the earliest the system shows, and I can't promise anything sooner. The only other opening is tomorrow morning, and that's a standard visit, which doesn't fit a burst pipe. Until the technician arrives, keep the water off, stay clear of the power strip, and leave the kitchen circuits off if you were able to flip them. If water starts coming from somewhere else, shut off the main valve. Is the service address still 1480 Larkspur Avenue, Unit 4B? And is 503-555-0193 the best callback number? Does the $99 fee work for you for the 2:00 to 4:00 PM window today? Also, is there a gate code or any access instructions the technician should

AGENThave?

CUSTOMERThe address is still 1480 Larkspur Avenue, but I moved to Unit 4D last month, not 4B. The callback number is right: 503-555-0193. The fee is fine for the 2 to 4 PM window, but if anything opens up earlier, please let me know. The gate code is pound 2719. Who will be coming, and how will I know they're on the way? Also, can you give me the ticket number for this visit?

AGENTThanks, I'll use Unit 4D. I'll open the ticket now and book the window.

CUSTOMEROkay. I'll wait for the details.

AGENTYou're booked. Here are the details: - Ticket number: CPL-58210 - Address: 1480 Larkspur Avenue, Unit 4D (I used 4D, not the 4B on file) - Gate code: pound 2719 - Callback number: 503-555-0193 - Window: today, 2:00 PM to 4:00 PM - Fee: $99 emergency callout (Copperline Care rate)

- TechnicianDana

Dana will text 503-555-0193 about 30 minutes before arriving, so keep your phone handy.

I can't see any earlier emergency openings right now, and I can't set up an alert if one appears. If you want a shot at an earlier slot, call us back and we'll check again.

Until Dana arrives, keep the water off and stay away from the power strip and the wet area. If you see sparks, smell burning, or water starts coming from somewhere else, shut off the main water valve or breaker and call us right away. If you ever feel in danger, call 911.

Is there anything else I can help

AGENTwith?

CUSTOMERNo, that covers it. Thanks for your help. Goodbye.

AGENTYou're welcome. Stay safe, and Dana will see you this afternoon. Goodbye.

Insurance

AGENTNorthgate

CUSTOMERSorry, um, could you say that again? I'm not sure what you mean by Northgate. Are you asking about my location, or did you need something else?

AGENTSorry for the confusion. My name is Alex, and I'm the claims assistant with Northgate Mutual Insurance. Northgate is just the name of the insurance company, and I'm not asking about your location. I'm here to help you report an auto accident. First, are you safe right now, and is anyone injured? If anyone is hurt or in danger, please hang up and call 911 first. If everyone is okay, I'll need your policy number and the last name of the policyholder so I can pull up your policy.

CUSTOMEROh, sorry. I'm a little flustered. Yeah, everyone's okay. Nobody's hurt. It's just me here with my baby, and he's kind of fussing, but we're all right. So my policy number is, uh, NG4471903. No, sorry, the last one is 2, and the last name is Carter. Is that okay?

AGENTThat's completely fine, take your time. I'm glad you and your baby are okay. I'll use the corrected number, NG4471902, and look it up now.

AGENTThank you,

CUSTOMERThanks.

AGENTThanks for your patience, Ms. Carter. I've verified your policy, NG4471902, for your 2021 Honda CR-V. Let me ask a few questions about what happened. 1. When did the accident happen (date and approximate time)? 2. Where did it happen? A street address or nearest intersection is fine. 3. Can you tell me what happened? Go at your own pace.

CUSTOMEROh, um, so this just happened about 20 minutes ago. Uh, today. I was stopped at a red light at Southeast 5th Avenue and Alder Street, and another car hit the back of mine. The back bumper's crushed, and the rear wheel's rubbing, so I can't drive it. Police didn't come out. Is that enough for you, or do you need more details?

AGENTThat's very helpful, thank you. Since the rear wheel is rubbing and the car isn't safe to drive, I can arrange a tow. I'll also check what your policy covers for towing and rental. First, I need the other driver's information:

1. Their name

2. Their insurance company

3. Their license plate number Also, is a phone number you can be reached at, so the tow driver can call you? And are you and the car still at Southeast 5th Avenue and Alder Street?

CUSTOMERUh, yeah. I'm still right here at Southeast 5th and Alder, standing next to the car. The other driver's name is Kevin Lund. His insurance is with Pacific Shore, and the plate number is 7KXR214. X like X-ray. My phone number is 660-386-1106. Sorry, is all that okay for you?

AGENTThat's all good, thank you. Let me get the tow and the claim started.

AGENTNow I'll

CUSTOMEROkay.

CUSTOMERAll right. Go ahead. Sorry, I'm just a little anxious. Please let me know what happens next.

AGENTNo need to apologize, you're doing fine. Everything is set up. Here's where things stand. Your claim - Claim number: NGM-904417 - Your adjuster is Priya Shah, and she will contact you within one business day. - Policy number: NG4471902 - Other driver: Kevin Lund, Pacific Shore, license plate 7KXR214 (X as in X-ray). Is that plate correct? Your tow - Eastside Towing has been dispatched to Southeast 5th Avenue and Alder Street. - The estimated arrival is about 45 minutes. - They will take your CR-V to our partner repair shop at 9th and Morrison. - The driver can reach you at 660-386-1106. - Your policy covers towing up to 25 miles. Your coverage - Collision deductible: $500. - Rental: $40 per day for up to 30 days, starting from the day the claim was opened. While you wait

- Stay with the car in a safe spot, away from traffic. If it's safe, take photos of the damage, the other car, and the intersection.

- Keep Kevin Lund's information handy.

- Since you have your baby with you, you may want to sit somewhere safe and comfortable while you wait.

Priya can answer questions about the repair and your rental. Do you have any questions for me, or is there anything

AGENTI got wrong?

CUSTOMERThank you for going through all that. I just want to make sure. So the claim number is NGM-904417, right? And, um, you said the deductible's 500 dollars, but I mean, he hit me while I was stopped, so I shouldn't have to pay that, right? And, uh, the rental car, that's covered for 40 a day, but is there any way to know if my premium's going to go up? I'm just worried about that. Also, can you tell me again where my car's being taken? Sorry, just want to make sure I got everything straight.

AGENTYes, the claim number is NGM-904417. Deductible: Your policy shows a $500 collision deductible. I can't say whether you will end up paying it. That depends on how the claim is handled and on who is found at fault, and I don't have that information. Being hit while stopped is something Priya Shah can look at when she reviews the claim, so please ask her about it. Rental: It's covered at $40 per day for up to 30 days, starting from the day the claim was opened. Premium: I don't have anything that tells me whether your premium could change, and I don't want to guess. Priya or your agent can answer that. Tow destination: Eastside Towing is taking your CR-V to our partner repair shop at 9th and Morrison. They should arrive in about 45 minutes, and the driver can reach you at 660-386-1106. I also never heard back on whether the plate I have for Kevin Lund is right. I have 7KXR214, with the X as in X-ray. Is

AGENTthat correct?

CUSTOMERYes, that's right. 7KXR214. X like X-ray. Sorry, I missed saying that before. Thank you for explaining everything. I just worry about all the details, so I really appreciate it. All right. I think I got what I need. Thank you so much, and goodbye.

AGENTYou're welcome, Ms. Carter, and you don't need to apologize. Asking about the details was the right thing to do. I've confirmed the plate as 7KXR214 (X as in X-ray). Your claim number is NGM-904417, and Priya Shah will contact you within one business day. Take care of yourself and your little one while you wait for the tow. Goodbye.

How these numbers are measured →