DATASETS2SCOVAL BENCHMARK ROTATIONACTIVE

s2s-workflow-bank-hard-v1 speech-to-speech benchmark dataset

s2s-workflow-bank-hard-v1 is a speech-to-speech test set in Coval's daily rotation — every measured speech-to-speech model runs the same fixed inputs from it.

Models measured
10
last 30 days
Current leader · V2V#1 / 10
789ms
Prismvia Rime

How models rank on s2s-workflow-bank-hard-v1

Full S2S dashboard
Voice-to-Voice Latency on s2s-workflow-bank-hard-v1Milliseconds · lower is better · 30-day average on this datasetEvery S2S model measured on the s2s-workflow-bank-hard-v1 dataset, ranked on Voice-to-Voice Latency.
median of all models · 1224 ms
S2S models on the s2s-workflow-bank-hard-v1 dataset over the last 30 days, ranked on Voice-to-Voice Latency.
#ModelHostV2VSamples
1PrismRime789 ms30
2Gemini 3.1 Flash Live (Preview)Google1041 ms140
3Gemini 3.8 LiveGoogle1050 ms141
4Dialog Rsn 1PolyAI1093 ms72
5Phonic 1Phonic1155 ms96
6Qwen3.8 Omni Flash RealtimeAlibaba1292 ms120
7GPT-Live-1OpenAI1399 ms89
8Eleven Realtime v1 MiniElevenLabs1627 ms120
9Eleven Realtime v1 MaxElevenLabs1674 ms120
10Grok Voice 3.0SpaceXAI1815 ms120
Under-sampled models are excluded; tied models share a place.

Metrics reported for this dataset

Same datasets, prompts and metric definitions for every model, measured by Coval’s open-source runner. Full methodology on the overview.

Evaluate your own voice agent

Use Coval to test your production configuration, prompts and calls—not only the public benchmark endpoints.

Book a Demo