# Talking Head Generation on VoxCeleb and LRW (test)

Source: [Talking Head Generation on VoxCeleb and LRW (test)](<https://www.sota2.com/research/sota/talking-head-generation-on-voxceleb-and-lrw-test>)

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

Data updated: 2026-02-26T05:48:37.976Z

Evaluation of Talking Head Generation models using the combined test splits of the VoxCeleb and LRW datasets, assessed via metrics like Lip Sync Quality, Head Movement Naturalness, and Video Realness.

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 | Talking Head Generation |
| Dataset (including any stated split) | VoxCeleb and LRW (test) |
| Metric | Lip Sync Quality |
| Unit | SCALAR |
| Direction | Higher 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

- [Lip Sync Quality](<https://www.sota2.com/research/sota/talking-head-generation-on-voxceleb-and-lrw-test.md?metric=Lip%20Sync%20Quality>): Higher is better; unit: SCALAR.

- [Head Movement Naturalness](<https://www.sota2.com/research/sota/talking-head-generation-on-voxceleb-and-lrw-test.md?metric=Head%20Movement%20Naturalness>): Higher is better; unit: SCALAR.

- [Video Realness](<https://www.sota2.com/research/sota/talking-head-generation-on-voxceleb-and-lrw-test.md?metric=Video%20Realness>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Lip Sync Quality | Result date | Report paper |
| --- | --- | --- | --- |
| Wav2Lip | 3.98 | Not recorded | 68fa69cb1e209131cea24114 |
| PC-AVS | 3.98 | Not recorded | 68fa69cb1e209131cea24114 |
| ATVG | 2.87 | Not recorded | 68fa69cb1e209131cea24114 |
| MakeitTalk | 2.33 | Not recorded | 68fa69cb1e209131cea24114 |

### Wav2Lip

Lip Sync Quality: 3.98. Result ID: 698de3d3ffc3901de09964e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea24114.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68361e209131cea1ad36.md>)

[Implementation](<https://www.sota2.com/github.com/Rudrabha/Wav2Lip>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PC-AVS

Lip Sync Quality: 3.98. Result ID: 698de3d3ffc3901de09964e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea24114.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea24114.md>)

[Implementation](<https://hangz-nju-cuhk.github.io/projects/PC-AVS>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ATVG

Lip Sync Quality: 2.87. Result ID: 698de3d3ffc3901de09964e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea24114.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65ba1e209131cea0be57.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MakeitTalk

Lip Sync Quality: 2.33. Result ID: 698de3d3ffc3901de09964e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea24114.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69531e209131cea2191a.md>)

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.