# Audio-visual Dialogue Generation on DIADG evaluation (mean of 1000 samples)

Source: [Audio-visual Dialogue Generation on DIADG evaluation (mean of 1000 samples)](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples>)

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

Data updated: 2026-07-07T07:39:47.590Z

Evaluation of audio-visual dialogue generation performance on the DIADG dataset, specifically using the mean result across 1000 samples, assessed using metrics like FID, FVD, LSE-C, and LPIPS.

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 dialogue generation |
| Dataset (including any stated split) | DIADG evaluation (mean of 1000 samples) |
| Metric | FID |
| Unit | SCALAR |
| Direction | Lower 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

- [FID](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=FID>): Lower is better; unit: SCALAR.

- [FVD](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=FVD>): Lower is better; unit: SCALAR.

- [LSE-C](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=LSE-C>): Higher is better; unit: SCALAR.

- [LSE-D](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=LSE-D>): Higher is better; unit: SCALAR.

- [FRCorr](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=FRCorr>): Higher is better; unit: SCALAR.

- [FRDiv](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=FRDiv>): Lower is better; unit: SCALAR.

- [CSIM](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=CSIM>): Higher is better; unit: SCALAR.

- [LPIPS](<https://www.sota2.com/research/sota/audio-visual-dialogue-generation-on-diadg-evaluation-mean-of-1000-samples.md?metric=LPIPS>): Lower is better; unit: PERCENT.

## Recorded results

| Method | FID | Result date | Report paper |
| --- | --- | --- | --- |
| CHAT | 17.33 | Not recorded | 6a4c87f985dba69fef8b2564 |
| EDTalk | 18.74 | Not recorded | 6a4c87f985dba69fef8b2564 |
| Hallo3 | 20.56 | Not recorded | 6a4c87f985dba69fef8b2564 |
| ReactDiff | 21.36 | Not recorded | 6a4c87f985dba69fef8b2564 |
| SadTalker | 22.53 | Not recorded | 6a4c87f985dba69fef8b2564 |
| DIM | 36.52 | Not recorded | 6a4c87f985dba69fef8b2564 |

### CHAT

FID: 17.33. Result ID: 6a4cad2760b88bb339c7b661.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

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

### EDTalk

FID: 18.74. Result ID: 6a4cad2760b88bb339c7b65f.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72c21e209131cea58724.md>)

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

### Hallo3

FID: 20.56. Result ID: 6a4cad2760b88bb339c7b65c.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa776a1e209131cea73230.md>)

[Implementation](<https://fudan-generative-vision.github.io/hallo3>)

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

### ReactDiff

FID: 21.36. Result ID: 6a4cad2760b88bb339c7b660.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7be11e209131cea8b000.md>)

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

### SadTalker

FID: 22.53. Result ID: 6a4cad2760b88bb339c7b65d.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ec81e209131cea420cd.md>)

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

### DIM

FID: 36.52. Result ID: 6a4cad2760b88bb339c7b65e.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87f985dba69fef8b2564.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74251e209131cea5ff64.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.