Sindhi Evaluation of Audio AI Models Sindhi Evaluation of Audio AI Models By Gooey.AI, CLEAR Global and the Gates Foundation • 28 Apr 2026 • Gooey.AI/language 0 5 10 15 20 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Accuracy Latency Lower is better More accurate GPT 5.5 — Accuracy 0.55, Latency 13.05GPT 5.5 GPT 5.5 + Omni — Accuracy 0.73, Latency 11.95GPT 5.5 + OmniMost Accurate Gemini 3.1 Pro — Accuracy 0.69, Latency 12.62Gemini 3.1 Pro Gemini 3.1 Pro + Omni — Accuracy 0.69, Latency 14.52Gemini 3.1 Pro + Omni Gemini 3.1 Flash-Lite — Accuracy 0.32, Latency 5.14Gemini 3.1 Flash-LiteFastest Gemini 3.1 Flash-Lite + Omni — Accuracy 0.55, Latency 7.28Gemini 3.1 Flash-Lite + Omni Kimi K2.6 + Omni — Accuracy 0.43, Latency 15.55Kimi K2.6 + Omni GPT-OSS 120B + Omni — Accuracy 0.35, Latency 8.8GPT-OSS 120B + Omni GPT-Realtime 1.5 — Accuracy 0.06, Latency 10.44GPT-Realtime 1.5 GPT-Realtime 1.5 + Omni — Accuracy 0.45, Latency 12.66GPT-Realtime 1.5 + Omni Gemma 4 26B + Omni — Accuracy 0.35, Latency 7.38Gemma 4 26B + Omni GPT 5.5 — Accuracy 0.55, Latency 13.050 GPT 5.5 + Omni — Accuracy 0.73, Latency 11.951 Gemini 3.1 Pro — Accuracy 0.69, Latency 12.622 Gemini 3.1 Pro + Omni — Accuracy 0.69, Latency 14.523 Gemini 3.1 Flash-Lite — Accuracy 0.32, Latency 5.144 Gemini 3.1 Flash-Lite + Omni — Accuracy 0.55, Latency 7.285 Kimi K2.6 + Omni — Accuracy 0.43, Latency 15.556 GPT-OSS 120B + Omni — Accuracy 0.35, Latency 8.87 GPT-Realtime 1.5 — Accuracy 0.06, Latency 10.448 GPT-Realtime 1.5 + Omni — Accuracy 0.45, Latency 12.669 Gemma 4 26B + Omni — Accuracy 0.35, Latency 7.3810 © 2026 Gooey.AI