# Speech Decoding on Brain-to-Text 2024 (hold-out) Source: [Speech Decoding on Brain-to-Text 2024 (hold-out)]() Collection index: [llms.txt]() Data updated: 2026-03-04T08:44:12.446Z Evaluation of Speech Decoding performance on the hold-out split of the Brain-to-Text 2024 dataset, measured using WER. 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 | Speech decoding | | Dataset (including any stated split) | Brain-to-Text 2024 (hold-out) | | Metric | WER | | Unit | PERCENT | | Direction | Lower is better | | Results loaded | 9 of 9 | | 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. ## Recorded results | Method | WER | Result date | Report paper | | --- | --- | --- | --- | | BIT | 5.1 | Not recorded | 692d20e5d6b0dd3342931455 | | Feghhi et al. (2025) | 5.68 | Not recorded | 692d20e5d6b0dd3342931455 | | Li et al. (2024) | 5.77 | Not recorded | 692d20e5d6b0dd3342931455 | | BIT | 6.35 | Not recorded | 692d20e5d6b0dd3342931455 | | Feghhi et al. (2025) | 7.98 | Not recorded | 692d20e5d6b0dd3342931455 | | RNN (Baseline) | 9.76 | Not recorded | 692d20e5d6b0dd3342931455 | | BIT | 10.22 | Not recorded | 692d20e5d6b0dd3342931455 | | BIT | 15.67 | Not recorded | 692d20e5d6b0dd3342931455 | | Feng et al. (2024) | 24.69 | Not recorded | 692d20e5d6b0dd3342931455 | ### BIT WER: 5.1. Result ID: 69a7f0a56dbb7075b94cd6b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding framework":"Cascaded","Ensemble":"true"}. This record does not assert independent reproduction. ### Feghhi et al. (2025) WER: 5.68. Result ID: 69a7f0a56dbb7075b94cd6b3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Ensemble":"true"}. This record does not assert independent reproduction. ### Li et al. (2024) WER: 5.77. Result ID: 69a7f0a56dbb7075b94cd6b2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Ensemble":"true"}. This record does not assert independent reproduction. ### BIT WER: 6.35. Result ID: 69a7f0a56dbb7075b94cd6b1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding framework":"Cascaded","Ensemble":"false"}. This record does not assert independent reproduction. ### Feghhi et al. (2025) WER: 7.98. Result ID: 69a7f0a56dbb7075b94cd6b0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Ensemble":"false"}. This record does not assert independent reproduction. ### RNN (Baseline) WER: 9.76. Result ID: 69a7f0a56dbb7075b94cd6af. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BIT WER: 10.22. Result ID: 69a7f0a56dbb7075b94cd6ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding framework":"End-to-End","Ensemble":"true"}. This record does not assert independent reproduction. ### BIT WER: 15.67. Result ID: 69a7f0a56dbb7075b94cd6ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Decoding framework":"End-to-End","Ensemble":"false"}. This record does not assert independent reproduction. ### Feng et al. (2024) WER: 24.69. Result ID: 69a7f0a56dbb7075b94cd6ac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.