# Ambient Sound Generation on VABench (test)

Source: [Ambient Sound Generation on VABench (test)](<https://www.sota2.com/research/sota/ambient-sound-generation-on-vabench-test>)

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

Data updated: 2026-05-13T10:44:35.838Z

Evaluation of models performing Ambient Sound Generation on the test split of the VABench dataset, assessed using metrics including DeSync, VA-IB, CLAP, and CLIP.

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 | Ambient Sound Generation |
| Dataset (including any stated split) | VABench (test) |
| Metric | DeSync |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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

- [DeSync](<https://www.sota2.com/research/sota/ambient-sound-generation-on-vabench-test.md?metric=DeSync>): Lower is better; unit: SCALAR.

- [VA-IB](<https://www.sota2.com/research/sota/ambient-sound-generation-on-vabench-test.md?metric=VA-IB>): Higher is better; unit: SCALAR.

- [CLAP](<https://www.sota2.com/research/sota/ambient-sound-generation-on-vabench-test.md?metric=CLAP>): Higher is better; unit: SCALAR.

- [CLIP](<https://www.sota2.com/research/sota/ambient-sound-generation-on-vabench-test.md?metric=CLIP>): Higher is better; unit: SCALAR.

## Recorded results

| Method | DeSync | Result date | Report paper |
| --- | --- | --- | --- |
| SyncDPO | 0.43 | Not recorded | 6a0405dbdbc836b87d827883 |
| SFT | 0.49 | Not recorded | 6a0405dbdbc836b87d827883 |
| SyncDPO | 0.51 | Not recorded | 6a0405dbdbc836b87d827883 |
| SFT | 0.54 | Not recorded | 6a0405dbdbc836b87d827883 |
| DPO | 0.67 | Not recorded | 6a0405dbdbc836b87d827883 |
| DPO | 0.68 | Not recorded | 6a0405dbdbc836b87d827883 |
| No Tune | 0.7 | Not recorded | 6a0405dbdbc836b87d827883 |

### SyncDPO

DeSync: 0.43. Result ID: 6a04555647f34ce924eaecf5.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

Recorded evaluation configuration: {"Training Data":"OOD"}. This record does not assert independent reproduction.

### SFT

DeSync: 0.49. Result ID: 6a04555647f34ce924eaecf3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bd21e209131cea8ab7e.md>)

Recorded evaluation configuration: {"Training Data":"OOD"}. This record does not assert independent reproduction.

### SyncDPO

DeSync: 0.51. Result ID: 6a04555647f34ce924eaecf2.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

Recorded evaluation configuration: {"Training Data":"ID"}. This record does not assert independent reproduction.

### SFT

DeSync: 0.54. Result ID: 6a04555647f34ce924eaecf0.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bd21e209131cea8ab7e.md>)

Recorded evaluation configuration: {"Training Data":"ID"}. This record does not assert independent reproduction.

### DPO

DeSync: 0.67. Result ID: 6a04555647f34ce924eaecf1.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Training Data":"ID"}. This record does not assert independent reproduction.

### DPO

DeSync: 0.68. Result ID: 6a04555647f34ce924eaecf4.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Training Data":"OOD"}. This record does not assert independent reproduction.

### No Tune

DeSync: 0.7. Result ID: 6a04555647f34ce924eaecef.

[Reporting paper](<https://www.sota2.com/research/paper/6a0405dbdbc836b87d827883.md>)

Recorded evaluation configuration: {"Training Data":"None"}. This record does not assert independent reproduction.

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