# Audio Deepfake Detection on ADD 2023 R2 (EER) Source: [Audio Deepfake Detection on ADD 2023 R2 (EER)]() Collection index: [llms.txt]() Data updated: 2026-03-09T08:10:41.923Z Evaluation of Audio Deepfake Detection performance using the ADD 2023 R2 dataset, specifically reporting the Equal Error Rate (EER). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Audio Deepfake Detection | | Dataset (including any stated split) | ADD 2023 R2 | | Metric | EER | | Unit | PERCENT | | Direction | Lower is better | | Results loaded | 19 of 19 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [EER](): Lower is better; unit: PERCENT. ## Recorded results | Method | EER | Result date | Report paper | | --- | --- | --- | --- | | DF-Arena 500M | 12.3 | Not recorded | 69ae53e32d603e6de3e428fe | | mHuBERT-Iter1 | 16.1 | Not recorded | 69ae53e32d603e6de3e428fe | | W2V2-TCM | 16.89 | Not recorded | 69ae53e32d603e6de3e428fe | | DF-Arena 100M-V1 | 17.04 | Not recorded | 69ae53e32d603e6de3e428fe | | W2V2-AASIST | 17.44 | Not recorded | 69ae53e32d603e6de3e428fe | | mHuBERT-Iter2 | 18.7 | Not recorded | 69ae53e32d603e6de3e428fe | | MoLEX | 19.02 | Not recorded | 69ae53e32d603e6de3e428fe | | WavLM-Base+ | 19.59 | Not recorded | 69ae53e32d603e6de3e428fe | | XLSR+Mamba | 20.15 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | mHuBERT-Final | 20.52 | Not recorded | 69ae53e32d603e6de3e428fe | | HuBERT-Base | 20.75 | Not recorded | 69ae53e32d603e6de3e428fe | | XLSR+SLS | 21.09 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | HyperPotter | 21.84 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | Wav2Vec2+AASIST | 21.92 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | Wav2Vec2+AASIST(Baseline) | 22.21 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | WavLM-Base | 22.38 | Not recorded | 69ae53e32d603e6de3e428fe | | XLSR+Conformer+TCM | 22.74 | Not recorded | 698575c7c97a0b3e3b5d5dbb | | ResembleAI-2B | 28.27 | Not recorded | 69ae53e32d603e6de3e428fe | | XLSR+BiCrossMamba | 29.93 | Not recorded | 698575c7c97a0b3e3b5d5dbb | ### DF-Arena 500M EER: 12.3. Result ID: 69ae806d6dbb7075b95683e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### mHuBERT-Iter1 EER: 16.1. Result ID: 69ae806d6dbb7075b95683eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### W2V2-TCM EER: 16.89. Result ID: 69ae806d6dbb7075b95683e5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### DF-Arena 100M-V1 EER: 17.04. Result ID: 69ae806d6dbb7075b95683e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### W2V2-AASIST EER: 17.44. Result ID: 69ae806d6dbb7075b95683e4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### mHuBERT-Iter2 EER: 18.7. Result ID: 69ae806d6dbb7075b95683ec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### MoLEX EER: 19.02. Result ID: 69ae806d6dbb7075b95683e7. [Reporting paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### WavLM-Base+ EER: 19.59. Result ID: 69ae806d6dbb7075b95683ef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### XLSR+Mamba EER: 20.15. Result ID: 6999fd1b5698665ec47ed29f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Params (M)":"319"}. This record does not assert independent reproduction. ### mHuBERT-Final EER: 20.52. Result ID: 69ae806d6dbb7075b95683ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### HuBERT-Base EER: 20.75. Result ID: 69ae806d6dbb7075b95683ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### XLSR+SLS EER: 21.09. Result ID: 6999fd1b5698665ec47ed29e. [Reporting paper]() Recorded evaluation configuration: {"Params (M)":"340"}. This record does not assert independent reproduction. ### HyperPotter EER: 21.84. Result ID: 6999fd1b5698665ec47ed2a2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params (M)":"317.87"}. This record does not assert independent reproduction. ### Wav2Vec2+AASIST EER: 21.92. Result ID: 6999fd1b5698665ec47ed29c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params (M)":"317.84"}. This record does not assert independent reproduction. ### Wav2Vec2+AASIST(Baseline) EER: 22.21. Result ID: 6999fd1b5698665ec47ed2a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params (M)":"317.84"}. This record does not assert independent reproduction. ### WavLM-Base EER: 22.38. Result ID: 69ae806d6dbb7075b95683ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### XLSR+Conformer+TCM EER: 22.74. Result ID: 6999fd1b5698665ec47ed29d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Params (M)":"319"}. This record does not assert independent reproduction. ### ResembleAI-2B EER: 28.27. Result ID: 69ae806d6dbb7075b95683e6. [Reporting paper]() Recorded evaluation configuration: {"Training Protocol":"Protocol 2 (multi-dataset)"}. This record does not assert independent reproduction. ### XLSR+BiCrossMamba EER: 29.93. Result ID: 6999fd1b5698665ec47ed2a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params (M)":"318.21"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.