# Audio-Visual Speech Recognition on MISP2022 (test)

Source: [Audio-Visual Speech Recognition on MISP2022 (test)](<https://www.sota2.com/research/sota/audio-visual-speech-recognition-on-misp2022-test>)

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

Data updated: 2026-02-26T06:09:12.318Z

Evaluation of Audio-Visual Speech Recognition performance on the test split of the MISP2022 dataset, measured using Character Error Rate (CER) and character-based Phoneme Error Rate (cpCER).

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-Visual Speech Recognition |
| Dataset (including any stated split) | MISP2022 (test) |
| Metric | cpCER |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 4 of 4 |
| 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

- [cpCER](<https://www.sota2.com/research/sota/audio-visual-speech-recognition-on-misp2022-test.md?metric=cpCER>): Lower is better; unit: PERCENT.

- [CER](<https://www.sota2.com/research/sota/audio-visual-speech-recognition-on-misp2022-test.md?metric=CER>): Lower is better; unit: PERCENT.

## Recorded results

| Method | cpCER | Result date | Report paper |
| --- | --- | --- | --- |
| MDA-KD | 28.06 | Not recorded | 68fa72791e209131cea5698d |
| NPU | 29.13 | Not recorded | 68fa72791e209131cea5698d |
| NIO | 29.58 | Not recorded | 68fa72791e209131cea5698d |
| XMU | 31.88 | Not recorded | 68fa72791e209131cea5698d |

### MDA-KD

cpCER: 28.06. Result ID: 69923b403a6c3148490d7dd9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72791e209131cea5698d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72791e209131cea5698d.md>)

[Implementation](<https://github.com/dalision/ModalBiasAVSR>)

Recorded evaluation configuration: {"Training Data (Audio)":"1000 hours","Training Data (Video)":"w/o extra data","Backbone":"Conformer","Objective Function":"ED + InterCTC"}. This record does not assert independent reproduction.

### NPU

cpCER: 29.13. Result ID: 69923b403a6c3148490d7dd8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72791e209131cea5698d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72cb1e209131cea58766.md>)

Recorded evaluation configuration: {"Training Data (Audio)":"1300 hours","Training Data (Video)":"w/o extra data","Backbone":"E-Branchformer","Objective Function":"ED + InterCTC"}. This record does not assert independent reproduction.

### NIO

cpCER: 29.58. Result ID: 69923b403a6c3148490d7dd6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72791e209131cea5698d.md>)

Recorded evaluation configuration: {"Training Data (Audio)":"3300 hours","Training Data (Video)":"LRW-1000","Backbone":"Conformer","Objective Function":"ED"}. This record does not assert independent reproduction.

### XMU

cpCER: 31.88. Result ID: 69923b403a6c3148490d7dd7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72791e209131cea5698d.md>)

Recorded evaluation configuration: {"Training Data (Audio)":"2100 hours","Training Data (Video)":"LRW-1000","Backbone":"Conformer","Objective Function":"ED + InterCTC"}. This record does not assert independent reproduction.

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