# ASR Error Correction on STOP (test) Source: [ASR Error Correction on STOP (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:08:29.111Z Evaluation of Automatic Speech Recognition (ASR) Error Correction performance on the test split of the STOP dataset, measured using WER and NE Hit Ratio. 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 | ASR error correction | | Dataset (including any stated split) | STOP (test) | | Metric | WER | | Unit | PERCENT | | Direction | Lower is better | | Results loaded | 18 of 18 | | 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 - [WER](): Lower is better; unit: PERCENT. - [NE Hit Ratio](): Higher is better; unit: PERCENT. ## Recorded results | Method | WER | Result date | Report paper | | --- | --- | --- | --- | | ORACLE | 5.7 | Not recorded | 68fa79451e209131cea7cfb6 | | ORACLE | 5.7 | Not recorded | 68fa79451e209131cea7cfb6 | | DeRAGEC | 5.8 | Not recorded | 68fa79451e209131cea7cfb6 | | DeRAGEC | 5.9 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.2 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.5 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.5 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.5 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.6 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 6.9 | Not recorded | 68fa79451e209131cea7cfb6 | | RAGEC | 7.2 | Not recorded | 68fa79451e209131cea7cfb6 | | GEC | 7.4 | Not recorded | 68fa79451e209131cea7cfb6 | | GEC | 7.8 | Not recorded | 68fa79451e209131cea7cfb6 | | ASR only | 8.9 | Not recorded | 68fa79451e209131cea7cfb6 | | ASR only | 8.9 | Not recorded | 68fa79451e209131cea7cfb6 | ### ORACLE WER: 5.7. Result ID: 6995d8863badfaffe6dea4ca. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Reference":"Ground truth of NE"}. This record does not assert independent reproduction. ### ORACLE WER: 5.7. Result ID: 6995d8863badfaffe6dea4cb. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Reference":"Ground truth of NE"}. This record does not assert independent reproduction. ### DeRAGEC WER: 5.8. Result ID: 6995d8863badfaffe6dea4c9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"MCQ + PS + Def + Rat"}. This record does not assert independent reproduction. ### DeRAGEC WER: 5.9. Result ID: 6995d8863badfaffe6dea4c8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"MCQ + PS + Def + Rat"}. This record does not assert independent reproduction. ### RAGEC WER: 6. Result ID: 6995d8863badfaffe6dea4c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"PS"}. This record does not assert independent reproduction. ### RAGEC WER: 6. Result ID: 6995d8863badfaffe6dea4c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"PS"}. This record does not assert independent reproduction. ### RAGEC WER: 6. Result ID: 6995d8863badfaffe6dea4c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"PS + Def"}. This record does not assert independent reproduction. ### RAGEC WER: 6.2. Result ID: 6995d8863badfaffe6dea4c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"PS + Def"}. This record does not assert independent reproduction. ### RAGEC WER: 6.5. Result ID: 6995d8863badfaffe6dea4be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"N (None)"}. This record does not assert independent reproduction. ### RAGEC WER: 6.5. Result ID: 6995d8863badfaffe6dea4c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"MCQ"}. This record does not assert independent reproduction. ### RAGEC WER: 6.5. Result ID: 6995d8863badfaffe6dea4c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"MCQ"}. This record does not assert independent reproduction. ### RAGEC WER: 6.6. Result ID: 6995d8863badfaffe6dea4bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"N (None)"}. This record does not assert independent reproduction. ### RAGEC WER: 6.9. Result ID: 6995d8863badfaffe6dea4c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini","Denoising feature":"Def"}. This record does not assert independent reproduction. ### RAGEC WER: 7.2. Result ID: 6995d8863badfaffe6dea4c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)","Denoising feature":"Def"}. This record does not assert independent reproduction. ### GEC WER: 7.4. Result ID: 6995d8863badfaffe6dea4bd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini"}. This record does not assert independent reproduction. ### GEC WER: 7.8. Result ID: 6995d8863badfaffe6dea4bc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)"}. This record does not assert independent reproduction. ### ASR only WER: 8.9. Result ID: 6995d8863badfaffe6dea4ba. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama-3.1 (70B)"}. This record does not assert independent reproduction. ### ASR only WER: 8.9. Result ID: 6995d8863badfaffe6dea4bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"gpt-4o-mini"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.