# Talking-head generation on Internal talking-head dataset In-domain (test)

Source: [Talking-head generation on Internal talking-head dataset In-domain (test)](<https://www.sota2.com/research/sota/talking-head-generation-on-internal-talking-head-dataset-in-domain-test>)

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

Data updated: 2026-03-04T09:14:19.456Z

Evaluation of Talking-head generation models on the in-domain test split of the internal talking-head dataset, using metrics like Lip-sync, Expressive, and Motion scores.

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) | Internal talking-head dataset In-domain (test) |
| Metric | Lip-sync Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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 Score](<https://www.sota2.com/research/sota/talking-head-generation-on-internal-talking-head-dataset-in-domain-test.md?metric=Lip-sync%20Score>): Higher is better; unit: SCALAR.

- [Expressive Score](<https://www.sota2.com/research/sota/talking-head-generation-on-internal-talking-head-dataset-in-domain-test.md?metric=Expressive%20Score>): Higher is better; unit: SCALAR.

- [Motion Score](<https://www.sota2.com/research/sota/talking-head-generation-on-internal-talking-head-dataset-in-domain-test.md?metric=Motion%20Score>): Higher is better; unit: SCALAR.

- [Average Score](<https://www.sota2.com/research/sota/talking-head-generation-on-internal-talking-head-dataset-in-domain-test.md?metric=Average%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Lip-sync Score | Result date | Report paper |
| --- | --- | --- | --- |
| RL | 4.42 | Not recorded | 69a66bde2d603e6de3e3ccd6 |
| Original | 4.03 | Not recorded | 69a66bde2d603e6de3e3ccd6 |
| SFT | 3.74 | Not recorded | 69a66bde2d603e6de3e3ccd6 |
| Sonic | 3.73 | Not recorded | 69a66bde2d603e6de3e3ccd6 |
| Memo | 3.53 | Not recorded | 69a66bde2d603e6de3e3ccd6 |
| Echo | 3.37 | Not recorded | 69a66bde2d603e6de3e3ccd6 |

### RL

Lip-sync Score: 4.42. Result ID: 69a7f7046dbb7075b94d01f2.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

Recorded evaluation configuration: {"training\_stage":"Reinforcement Learning"}. This record does not assert independent reproduction.

### Original

Lip-sync Score: 4.03. Result ID: 69a7f7046dbb7075b94d01ed.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

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

### SFT

Lip-sync Score: 3.74. Result ID: 69a7f7046dbb7075b94d01f1.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

Recorded evaluation configuration: {"training\_stage":"Supervised Fine-Tuning"}. This record does not assert independent reproduction.

### Sonic

Lip-sync Score: 3.73. Result ID: 69a7f7046dbb7075b94d01f0.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79311e209131cea7c8ac.md>)

[Implementation](<https://jixiaozhong.github.io/Sonic/>)

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

### Memo

Lip-sync Score: 3.53. Result ID: 69a7f7046dbb7075b94d01ef.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760f1e209131cea6b336.md>)

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

### Echo

Lip-sync Score: 3.37. Result ID: 69a7f7046dbb7075b94d01ee.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bde2d603e6de3e3ccd6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740d1e209131cea5f79d.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.
