# Text-to-Audio Generation on RiTTA Count OOD (test)

Source: [Text-to-Audio Generation on RiTTA Count OOD (test)](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test>)

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

Data updated: 2026-02-26T05:47:45.139Z

Evaluation of Text-to-Audio Generation models on the test split of the RiTTA Count OOD dataset, assessed using fidelity and quality metrics including KL divergence, FAD, FD, IS, and CLAP 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 | Text-to-Audio Generation |
| Dataset (including any stated split) | RiTTA Count OOD (test) |
| Metric | KL Divergence |
| Unit | SCALAR |
| 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

- [KL Divergence](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test.md?metric=KL%20Divergence>): Lower is better; unit: SCALAR.

- [FD (Fréchet Distance)](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test.md?metric=FD%20(Fr%C3%A9chet%20Distance)>): Lower is better; unit: SCALAR.

- [FAD (Fréchet Audio Distance)](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test.md?metric=FAD%20(Fr%C3%A9chet%20Audio%20Distance)>): Lower is better; unit: SCALAR.

- [IS (Inception Score)](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test.md?metric=IS%20(Inception%20Score)>): Higher is better; unit: SCALAR.

- [CLAP Score](<https://www.sota2.com/research/sota/text-to-audio-generation-on-ritta-count-ood-test.md?metric=CLAP%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | KL Divergence | Result date | Report paper |
| --- | --- | --- | --- |
| TangoFlux-RAG | 2.18 | Not recorded | 6909886f3ef7ccd74e5e5999 |
| TangoFlux | 2.22 | Not recorded | 6909886f3ef7ccd74e5e5999 |
| AudioLDM2-RAG | 2.71 | Not recorded | 6909886f3ef7ccd74e5e5999 |
| AudioLDM2 | 2.81 | Not recorded | 6909886f3ef7ccd74e5e5999 |

### TangoFlux-RAG

KL Divergence: 2.18. Result ID: 699adbb15698665ec46b5c9f.

[Reporting paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

Recorded evaluation configuration: {"base\_model":"TangoFlux"}. This record does not assert independent reproduction.

### TangoFlux

KL Divergence: 2.22. Result ID: 699adbb15698665ec46b5c9e.

[Reporting paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ed1e209131cea7613a.md>)

[Implementation](<https://github.com/declare-lab/TangoFlux>)

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

### AudioLDM2-RAG

KL Divergence: 2.71. Result ID: 699adbb15698665ec46b5c9d.

[Reporting paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

Recorded evaluation configuration: {"base\_model":"AudioLDM2"}. This record does not assert independent reproduction.

### AudioLDM2

KL Divergence: 2.81. Result ID: 699adbb15698665ec46b5c9c.

[Reporting paper](<https://www.sota2.com/research/paper/6909886f3ef7ccd74e5e5999.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73261e209131cea5ab21.md>)

[Implementation](<https://audioldm.github.io/audioldm2>)

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.