# Utterance-level Audio Spoofing Detection on LlamaPartialSpoof All spoof combined

Source: [Utterance-level Audio Spoofing Detection on LlamaPartialSpoof All spoof combined](<https://www.sota2.com/research/sota/utterance-level-audio-spoofing-detection-on-llamapartialspoof-all-spoof-combined>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-02T10:28:42.792Z

Evaluation of Utterance-level Audio Spoofing Detection performance using the combined set of all spoof types from the LlamaPartialSpoof dataset, reporting EER and AUC.

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 | Utterance-level Audio Spoofing Detection |
| Dataset (including any stated split) | LlamaPartialSpoof All spoof combined |
| Metric | EER |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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](<https://www.sota2.com/research/sota/utterance-level-audio-spoofing-detection-on-llamapartialspoof-all-spoof-combined.md?metric=EER>): Lower is better; unit: SCALAR.

- [AUC](<https://www.sota2.com/research/sota/utterance-level-audio-spoofing-detection-on-llamapartialspoof-all-spoof-combined.md?metric=AUC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | EER | Result date | Report paper |
| --- | --- | --- | --- |
| TRACE | 0.2208 | Not recorded | 69cdf8210636e5bf4afccba4 |
| TRACE | 0.2412 | Not recorded | 69cdf8210636e5bf4afccba4 |
| PartialSpoof-trained baseline | 0.2449 | Not recorded | 69cdf8210636e5bf4afccba4 |
| w2v2-XLsr-SAL | 0.3552 | Not recorded | 69cdf8210636e5bf4afccba4 |
| WavLM-SAL | 0.366 | Not recorded | 69cdf8210636e5bf4afccba4 |
| WavLM-BAM | 0.4258 | Not recorded | 69cdf8210636e5bf4afccba4 |
| w2v2-Large-MultiRes | 0.4749 | Not recorded | 69cdf8210636e5bf4afccba4 |

### TRACE

EER: 0.2208. Result ID: 69ce44cd47f34ce924b8219c.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"PartialSpoof-dev-calibrated","corpus":"Full Corpus"}. This record does not assert independent reproduction.

### TRACE

EER: 0.2412. Result ID: 69ce44cd47f34ce924b8219b.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal","corpus":"Full Corpus"}. This record does not assert independent reproduction.

### PartialSpoof-trained baseline

EER: 0.2449. Result ID: 69ce44cd47f34ce924b8219a.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767c1e209131cea6d579.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal"}. This record does not assert independent reproduction.

### w2v2-XLsr-SAL

EER: 0.3552. Result ID: 69ce44cd47f34ce924b82198.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697c3b6ea8b01e76cc8da749.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal"}. This record does not assert independent reproduction.

### WavLM-SAL

EER: 0.366. Result ID: 69ce44cd47f34ce924b82199.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697c3b6ea8b01e76cc8da749.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal"}. This record does not assert independent reproduction.

### WavLM-BAM

EER: 0.4258. Result ID: 69ce44cd47f34ce924b82197.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa747e1e209131cea61ee4.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal"}. This record does not assert independent reproduction.

### w2v2-Large-MultiRes

EER: 0.4749. Result ID: 69ce44cd47f34ce924b82196.

[Reporting paper](<https://www.sota2.com/research/paper/69cdf8210636e5bf4afccba4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7e1e209131cea404b2.md>)

Recorded evaluation configuration: {"Threshold calibration strategy":"subset-optimal"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.