List your company

Speech Translation on CoVoST-2 X-En (test)

43.4De PerformanceAudioPaLM2
716.4525.935.35May 19, 2023Nov 11, 2023May 6, 2024Oct 30, 2024Apr 24, 2025Oct 18, 2025Apr 13, 2026

Evaluation results

MethodLinks
AudioPaLM2Architecture Category=End-to-end ST Methods2024.0243.444.844.238.4—55.625.539.148.341307.648.73542.625.956.2
GenTranslate-V2Architecture Category=Cascaded ASR+MT Methods2024.0241.143.143.339.5—55.624.538.244.937.427.810.348.730.4422658.4
GenTranslate-V2Architecture Category=End-to-end ST Methods2024.0240.642.743.639.4—5423.337.744.83727.710.24830.542.325.455.9
SeamlessM4T-V2 (ASR+MT)Architecture Category=Cascaded ASR+MT Methods2024.0240.6434338.8—55.222.936.743.233.927.28.64727.841.924.753.1
GenTranslateArchitecture Category=Cascaded ASR+MT Methods2024.0240.241.842.138.4—53.722.936.943.834.329.49.549.731.239.622.354.6
SeamlessM4T-Large-V2Architecture Category=End-to-end ST Methods2024.024042.442.939—53.622.436.242.733.226.98.646.527.541.723.752.6
SeamlessM4T (ASR+MT)Architecture Category=Cascaded ASR+MT Methods2024.0239.841.541.137.5—53.221.435.142.429.926.5845.228.838.62250.6
GenTranslateArchitecture Category=End-to-end ST Methods2024.0239.241.74238.7—50.121.636.142.133.528.28.749.730.338.222.954.3
SeamlessM4T-LargeArchitecture Category=End-to-end ST Methods2024.0238.841.341.138.4—48.620.934.541.131.226.37.54526.537.621.851.4
Phi-4-mmLang.=X -> EN2026.0437.9—————24.9—————5.5——33.3—
Whisper + mBART-50Architecture Category=Cascaded ASR+MT Methods2024.023738.840.733—4921.530.739.932.716.30.43721.4252345.5
ComSL-LargeArchitecture Category=End-to-end ST Methods2024.023638.840.435.3—49.221.431.939.733.619.22.941.421.331.621.346.6
Whisper + NLLB-3.3bArchitecture Category=Cascaded ASR+MT Methods2024.0235.534.437.931.7—45.41929.839.826.717.50.13720.629.425.545.9
Whisper-Large V2Architecture Category=End-to-end ST Methods2024.023535.539.631—42.316.928.140.227.5140.238.51316.324.747.3
XLS-R-2bArchitecture Category=End-to-end ST Methods2024.0233.637.639.233.8—39.59.42231.716.711.11.617.119.519.63.516.5
AF-Next-instructLang.=X -> EN2026.0433—————25.6—————29.4——27.2—
Cascaded STAux. Data ASR=true, Aux. Data Text=None2023.052127.627.421.313.516.8719.2—————————
U2TT (LARGE) w/ DUBAux. Data ASR=true, Aux. Data Text=1M2023.0519.529.530.925.223.923.26.122.6—————————
Transformer-ST + ASR pre-trainAux. Data ASR=true, Aux. Data Text=None2023.0517.126.32318.811.314.85.816.7—————————
U2TT (LARGE)Aux. Data ASR=true, Aux. Data Text=None2023.0516.727.428.123.12021.95.920.5—————————
Revisit STAux. Data ASR=false, Aux. Data Text=None2023.0514.126.915.717.22.43.6211.7—————————
Transformer-STAux. Data ASR=false, Aux. Data Text=None2023.058.44.31214.40.21.21.48.8—————————