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

Scribe v3 Live + Claude Haiku 5.5 + v4 Turbo

Cascade20 scored calls · Nov 15, 2026

Rank on Roark score

#50of 929 stacks

60 / 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 price56 / 100323 of 929LeaderSoniox v6 + Muse Ember 1.0 + TTS-181 / 100
Task completion82%887 of 932LeaderLinden-1 (own VAD) + GPT-6.1 Sol + TTS-1 HD99%
Tool use69%334 of 908LeaderAssemblyAI-STT-3.7 Realtime-Agent + GPT-6.2 Sol + Flash v2.5100%
Time to first word0.7 s125 of 932LeaderUniversal-3.5 Pro + GPT-6.1 Sol + Flash v2.50.5 s
Response time2.8 s173 of 929LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Barge-in handlingn/aLeaderUniversal-3.6 Pro + Gemini 4 Neon + TTS-1100%
Turn budget2.8 s144 of 858LeaderLinden-1 (own VAD) + Claude Haiku 4.5 + TTS-11.2 s
Cost per task$0.66867 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
82%Completed

#1097 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
4.1/ 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
4.3/ 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.

Not scored by this sweep.

Incorrect interruptions

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

20 calls
0.00%avg

p50

0.00%

p95

0.00%

#1 of 1184 on the board · best 0%

Overtalk

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

20 calls
1.46%avg

p50

0.00%

p95

6.77%

#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.

31 calls
0.71per callavg

p50

1.00 per call

p95

1.00 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
0.66%avg

p50

0.24%

p95

2.44%

#301 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
1.86%avg

p50

1.49%

p95

4.26%

#192 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.

18 calls
92.8%avg

p50

98.3%

p95

100%

#201 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.

18 calls
1.94%avg

p50

1.28%

p95

4.53%

#46 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

#814 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.

145 turns
2.89savg

p50

2.77s

p95

4.23s

#255 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
0.79savg

p50

0.70s

p95

1.30s

#150 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
2.55per callavg

p50

2.00 per call

p95

6.05 per call

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

By industry

IndustryTask completionLatency P50Barge-in handling
Healthcare76%2.45 s—
Home services87%2.92 s—
Insurance74%2.60 s—
Customer support91%3.18 s—

By caller

CallerTask completionLatency P50Barge-in handling
US English, quiet line85%2.75 s—
Spanish-accented English, city street92%2.74 s—
Indian English, busy café92%2.56 s—
Southern US, flustered, crying baby92%2.85 s—
Fast and interrupting, construction outside48%2.91 s—

The calls behind the numbers

Customer support

AGENTBrightwave Internet

CUSTOMERHey. Uh, h-hi there.

AGENTHi, this is Alex with Brightwave Internet. I'm here, and I can help with your connection. Could you give me your account number or the phone number on the account?

CUSTOMEROh, uh, yeah. Let me grab that for you. Sorry. The baby's just... Anyway, my account number should be BW883017. Oh, shoot. Sorry. Last digit- One. Can you check if that's right on your end?

CUSTOMERHello?

AGENTNo problem, take your time. I'll look up that account now.

AGENTI found the account, BW883011. Let me check whether there's an outage in your area.

AGENTThanks for your patience. Good news: there's no known outage in your area. I'm going to run a line test on your connection now.

AGENTThe line test shows

CUSTOMEROkay. Um, sorry. Could you repeat that? I didn't catch the last part. Just got distracted for a second.

AGENTNo problem. Here's where things stand. Your account is BW883011. The line test showed a good signal to your gateway, but the gateway wasn't registered. I sent a remote reset, and it's now registered and online. Your devices may take about two minutes to reconnect. Can you check whether your internet is working now?

CUSTOMERAll right. Hang on a sec. Okay. The internet light just went solid white. Is that a good sign? How long till every- gets back online? And, uh, could I get a credit for the downtime? I lost two hours of work today.

AGENTA solid white light is a good sign. It means the gateway is registered and online. Your devices should reconnect within about two minutes. If anything is still offline after that, let me know. For the credit, I can apply one for the downtime. Should I go ahead with that on account BW883011?

CUSTOMERYes, please. Uh, sorry. Can you confirm that's enough to cover two hours? I really need it. Lost a lot of work. Is that okay?

AGENTI can apply a service credit for the downtime, but I can't tell you the exact dollar amount or whether it covers your lost work. The credit is applied to the account and the amount is set by our billing system. I'm applying it now on BW883011 for the downtime you had today.

AGENTA $10.00 service credit has been applied to account BW883011. It will show on your next bill. I know that may not match the work you lost. If you'd like, I can note your concern about the lost work, but I can't change the credit amount myself. Is your connection working now?

CUSTOMERYeah. It looks like everything's coming back online. Thanks. But, uh, I lost two hours. Could you maybe push for a bit more than 10 dollars? I'd really appreciate it if you can note that down.

AGENTI'm glad everything is coming back online. I can't change the $10.00 credit or push for a larger amount. The credit amount is set by billing, and I don't have a way to adjust it. I'm sorry this cost you two hours of work. I'll confirm the details: account BW883011, $10.00 credit applied for today's downtime, showing on your next bill. If you want to dispute the amount or ask for more, you can contact Brightwave billing directly, and they can review it with you. Is there anything else I can help with?

CUSTOMERNo. I think that should do it. Thank you for your help. I'll reach out to billing if I need to. Have a good day.

AGENTYou're welcome. To recap: account BW883011, gateway reset and online, and a $10.00 credit on your next bill. Have a good day, and I hope the rest of your work goes smoothly.

Healthcare

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

CUSTOMERHey, uh, hi, Alex. Sorry, I've got a baby fussing here, so I might sound a little distracted. I need to reschedule my dermatology appointment that's set for next Wednesday morning. I just got put on a work shift for that time, and I can only do after 3:00 p.m. now. Is that gonna be possible?

AGENTNo problem at all, I can help with that. First I need to verify your identity. Could you please give me your full name and date of birth?

CUSTOMEROh, sure. Um, it's Wade Carter, and my date of birth is March 17th, 1948. Oh, sorry. No, I misspoke. It's 1984. March 17th, 1984. Is that okay?

AGENTThank you, Wade. I'll verify that with 1984 as your birth year.

AGENTThanks, Wade, you're verified. Let me pull up your upcoming appointments.

AGENTI found your dermatology appointment with Dr. Okafor next Wednesday at 9:15 a.m. at Main Campus, Building B. What day and time after 3:00 p.m. would you like to move it to? If you have a specific day in mind, I'll check Dr. Okafor's availability for it, or I can look across the coming days.

CUSTOMEROkay. Um, thanks for- Second. Sorry again for the trouble. I can't do mornings now. Just after 3:00 p.m. Could you see what's available after 3:00 any day next week? If you've got something Thursday or Friday, that'd be best.

AGENTNo trouble at all. Let me check Dr. Okafor's availability.

AGENTHere's what Dr. Okafor has open that fits your constraints (after 3:00 p.m.): - Tuesday at 3:45

CUSTOMERHello? Oh, um, actually, sorry. Tuesdays won't work for me after all. I have to pick up my daughter on Tuesdays. D-do you have anything on Friday or, uh, maybe Thursday?

AGENTThursday at 4:10 p.m. works. Friday only has an 11:30 a.m. opening, which is before 3:00, so it doesn't fit your schedule. Would you like me to move your appointment to Thursday at 4:10 p.m. with Dr. Okafor?

CUSTOMERYes, uh, that should be fine. Thursday at 4:10 p.m., please. Sorry, is that gonna be a problem? Could you confirm that's all set?

CUSTOMERHello?

AGENTYes, Wade, that's no problem at all. I'm rescheduling your dermatology appointment with Dr. Okafor to Thursday at 4:10 p.m.

AGENTYour appointment is all set. Here are the details: - Provider: Dr. Okafor, Dermatology - New time: Thursday at 4:10 p.m. - Location: Main Campus, Building B - Confirmation code: RVB-3M8P2

Please arrive 15 minutes early with your photo ID and insurance card.

Is there anything else I can help you with?

CUSTOMERThank you. Uh, sorry. Just a couple quick questions. Is there any fee for changing the appointment like this?

CUSTOMERHello?

AGENTI'll check the clinic's change policy for you.

AGENTThere's no

CUSTOMEROkay.

AGENTSorry,

CUSTOMEROh, good. That's No, that's okay. I just wanted to make sure. Thank you for checking. Um, where should I park when I come in? I always get a bit turned around there.

AGENTGood news on the fee: since you're changing more than 24 hours before your appointment, there's no charge for this change. For parking, there's free patient parking in Lot C, with the entrance on River Road. Is there anything else I can help you with?

CUSTOMERThat's a relief. Thank you. Uh, last thing. Does Dr. Okafor accept Aetna insurance? I just wanna be sure before I show up.

AGENTI'm sorry, Wade, I don't have information about which insurance plans Dr. Okafor accepts. I can only help with scheduling, clinic hours, parking, and the change and cancellation policy. I'd suggest calling the clinic's front desk or checking with your insurance provider directly before your Thursday appointment. Is there anything else I can help with?

CUSTOMERNo. I think that's everything. Thanks for all your help. Uh, and just so I've got it, the new time is Thursday at 4:10 p.m., confirmation code RVB38- M8P2, right?

CUSTOMERHello ?

AGENTAlmost, Wade. The confirmation code is RVB-3M8P2. Your new appointment is Thursday at 4:10 p.m. with Dr. Okafor at Main Campus, Building B. Take care, and I hope your little one settles down soon.

Home services

AGENTCopperline Home Services,

CUSTOMERUh, hi. Sorry. I'm a little flustered. My baby's crying, and, uh, I've got a pipe under my kitchen sink that just burst. There's water all over the floor, and it's heading for a power strip. I, um, I need someone to come out as soon as possible. Can y'all help with emergencies like that?

AGENTYes, we can help. First, let's deal with the water and the power right now. 1. If you can safely reach it, shut off the water supply valves under the sink by turning them clockwise. If you can't reach them, or they don't turn, shut off the main water valve for the house. 2. Unplug the power strip and move it, along with anything else electrical, away from the water. Don't step into standing water near any outlets or appliances. 3. If the water is spreading or you think it's near an outlet, switch off the breaker for that kitchen area.

CUSTOMEROkay. Uh, hang on. L-let me try to turn the valve. Yeah. Okay. I turned it off. The water stopped. And I just unplugged the power strip, moved it out of the way. Sorry. I'm just a little scattered. I think nothing else electrical is close, so, um, that part should be safe now. What do I need to do for someone to come out? And, uh, can you tell me how much it's gonna cost before I go ahead?

AGENTThat's great, you did exactly the right things. Keep the baby and yourself away from the wet area until the technician arrives. To get someone out to you, I'll need to find your account first. What phone number do you have on file, or what's the service address?

CUSTOMEROh, thanks for saying that. I just worry I messed up. So my address is 1480 Larkspur Avenue, and the phone number should be 503-555-0139. No, 0193. Can you check that for me? I moved last month, so I'm in unit 4D now, not 4B. Is that all you need?

AGENTLet me look that up with the address and the number you gave me.

CUSTOMERAll right. Thank you. I appreciate it. Just let me know if you need anything else from me.

AGENTBefore we book, I still need to find your account. I haven't looked up your customer record yet, and I'm sorry about the delay. I'll run the lookup now using 1480 Larkspur Avenue, unit 4D, and 503-555-0193.

AGENTI found

CUSTOMEROkay. That's right.

CUSTOMERSorry. Was there more you needed to ask? I just wanna make sure I'm not missing anything.

CUSTOMERHello?

AGENTSorry, I was cut off there. I found your account, but it still lists unit 4B. You told me you moved to 4D, so I'll use 4D for this visit. Your account also shows a Copperline Care membership, which affects the fee. Here's what's available for an emergency visit: - Today, 2:00 PM to 4:00 PM (emergency visit) - Tomorrow, 8:00 AM to 10:00 AM (standard visit) For your membership, the emergency callout fee is $99. A standard visit is $79, but since the water is already off, an emergency visit is the right choice for today.

Does today at 200 PM to 4:00 PM work for you, and do you agree to the $99 fee? Once you confirm, I'll open the ticket and book the window. Please also tell me if there is a gate code or any access instructions for the building.

CUSTOMERThank you for checking. Uh, yeah, I moved to 4D, not 4B, so please use 4D. I'm glad you caught that. Um, is there any way someone can come before 2:00 p.m.? I know you said the water's off- But I'm just really worried about the mess and the baby. If that's the soonest, I'll take it, but I just wanted to ask. And, uh, the gate code is #2719. Will the $99 fee cover everything, or could there be extra charges?

AGENTI understand your concern. The soonest window the system shows for an emergency visit is today, 2:00 PM to 4:00 PM. I don't have any earlier openings to offer. About the fee: the $99 is the emergency callout fee for your Copperline Care membership. I

CUSTOMERAll right. Thank you for explaining. I just get anxious with this kind of thing, you know? So I'll take the 2:00 to 4:00 p.m. window, and I understand about the $99 fee. Can you tell me who's coming out and how I'll- ... when they're on the way? And, uh, will I get a ticket number or anything so I can check on the visit?

AGENTThanks for confirming. I'm opening the ticket now with the 4D unit and gate code, and then I'll book the 2:00 to 4:00 PM window.

CUSTOMEROkay. That sounds good. Just let me know once you have the ticket number and who will be coming, please.

AGENTYour visit is booked. - Ticket number: CPL-58210 - Technician: Dana - Window: Today, 2:00 PM to 4:00 PM

- Service address1480 Larkspur Avenue, Unit 4D

- Gate code#2719

- Callback number503-555-0193

- Fee$99 emergency callout (Copperline Care membership)

Dana will text the callback number about 30 minutes before arrival. Keep the water shut off, and stay away from the wet area and the outlets until then.

Is there anything else I can help you with?

CUSTOMERNo, that's everything. Thank you so much for your help today. I really appreciate it. Goodbye.

AGENTYou're welcome. I hope the baby settles soon. Goodbye.

Insurance

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

CUSTOMERUh, yeah, I think so. Sorry, I'm at home. My baby's crying, so I'm a little flustered. Nobody's hurt. I just-- I need some help with my car after an accident.

AGENTI'm glad you're safe and that nobody's hurt. Take a breath. I'll walk you through this step by step. To start, I need your policy number. If you know the policyholder's last name, please share that too.

CUSTOMEROkay. Thank you. I appreciate it. The policy number should be NG4471903. Oh, wait. Sorry. Last one is two. So NG4471902. And the last name is Carter. Is that all right?

CUSTOMERHello?

AGENTYes, I have that. Thanks for correcting it. Let me verify the policy now.

AGENTThank you, Ms. Carter. Your policy NG4471902 is verified, covering a 2021 Honda CR-V. Next, when and where did the accident happen? If you're not sure of the exact time, a rough estimate is fine.

CUSTOMERThanks, but it's actually Mr. Carter, Wade Carter. Sorry, just wanted to mention that. Accident happened about 20 minutes ago at the intersection of Southeast Fifth Avenue and Alder Street. I was stopped at a red light when it happened. Is that enough detail?

CUSTOMERHello?

CUSTOMERAre you still there?

CUSTOMERHello?

AGENTGot it, Mr. Carter. I'm sorry for the mix-up, and I'll note Wade Carter as the policyholder. Yes, that's a good start. The accident was about 20 minutes ago at Southeast Fifth Avenue and Alder Street, and you were stopped at a red light. I'm still here. Next, could you tell me what happened, meaning how the other vehicle hit you? Then I need the other driver's name, their insurance company, and their license plate number. Also, did the police come to the scene, and if so, do you have a report number?

CUSTOMERThank you. I appreciate you double-checking everything. So, uh, the other car hit the back of my car while I was stopped at the light. I didn't move, just sitting still, and he ran right into me. The other driver's name is Kevin Lund. His insurance is Pacific Shore, and the plate is 7K-X, like X-ray, R214. No police came out, so I don't have a report number. Is that going to be a problem?

AGENTNo, that's not a problem. A missing police report won't stop us from opening your claim. Let me confirm what I have: the other driver is Kevin Lund, insured with Pacific Shore, plate 7K-X-R214. Is that correct? Before I open the claim, can you tell me whether your car can be driven safely? If it's damaged or hard to drive, I can arrange a tow.

CUSTOMERYeah, that's right. You got all that correct. Uh, the back bumper's crushed and the rear wheel is rubbing real bad, so I can't drive it. I'd need a tow for sure. Where's it gonna be taken? And, um, am I gonna have to pay anything for this?

AGENTI'll arrange the tow and open the claim now. I also need to check your coverage so I can answer your cost question accurately.

AGENTHere's what your policy shows: your collision deductible is $500. Towing is covered up to 25 miles. Rental reimbursement is $40 per day for up to 30 days, starting the day the claim is opened. Now I'll open the claim and arrange the tow.

AGENTYour

CUSTOMERWait, but-

CUSTOMERSorry, just to check, he hit me, so I shouldn't have to pay the deductible, right? That's making me a little anxious. And, um, does the rental car start right away since I can't drive mine? Is my premium going to go up because of this?

AGENTI understand the worry. Here's what I can tell you from your policy, and what I can't. Your claim number is NGM-904417, and your adjuster is Priya Shah, who will contact you within one business day. The tow is dispatched with Eastside Towing, ETA about 45 minutes, taking your car to a Northgate partner repair shop at 9th and Morrison. On the deductible: I can't tell you whether it will be waived because the other driver was at fault. That's a decision for your adjuster once they review the claim. Your policy lists a $500 collision deductible, and the claim will be handled under that. On the rental: your policy covers $40 per day for up to 30 days, starting the day the claim is opened. I can't confirm whether a rental car can be set up immediately, so please ask Priya

CUSTOMEROkay. That helps a little. Thank you for explaining everything. Just so I'm clear, the claim number is NGM904417, right? And the tow is coming. Car's going to Knight & Morris. I'll wait for Priya to reach out about the deductible and the rental. Um, do you know if my premium is going to go up after this, or is that something only the adjuster can answer too?

CUSTOMERHello?

AGENTYes, the claim number is NGM-904417. One correction on the repair shop: the tow company is taking your car to a Northgate partner repair shop at 9th and Morrison, not Knight & Morris. Please confirm that with the tow driver when they arrive. As for your premium, I don't have any information about that, and I can't predict it. Your adjuster or your policy agent would be the right people to ask. Is there anything else you need right now, or anything about the scene you want to tell me before you wait for the tow?

CUSTOMERThank you for clarifying. So it's 9th and Morrison, not Knight and Morris. I'll make sure to double-check with the tow driver. I think that covers everything for now. I'm just anxious, but I really appreciate your help. I'll wait for the adjuster and the tow. Thank you again, and goodbye.

CUSTOMERHello?

AGENTYou're welcome, Mr. Carter. To recap: your claim number is NGM-904417, the tow is coming from Eastside Towing to the partner shop at 9th and Morrison, and Priya Shah will contact you within one business

How these numbers are measured →