# Text-to-Audio on RiTTA (test)

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

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

Data updated: 2026-04-10T06:48:31.877Z

Evaluation of Text-to-Audio models on the test split of the RiTTA dataset, assessed using metrics including FAD, KL divergence, FD, and various precision/recall metrics (mAPre, mARel, mAPar, mAMSR).

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 |
| Dataset (including any stated split) | RiTTA (test) |
| Metric | FAD |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 11 of 11 |
| 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

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

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

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

- [mAPre](<https://www.sota2.com/research/sota/text-to-audio-on-ritta-test.md?metric=mAPre>): Higher is better; unit: SCALAR.

- [mARel](<https://www.sota2.com/research/sota/text-to-audio-on-ritta-test.md?metric=mARel>): Higher is better; unit: SCALAR.

- [mAPar](<https://www.sota2.com/research/sota/text-to-audio-on-ritta-test.md?metric=mAPar>): Higher is better; unit: SCALAR.

- [mAMSR](<https://www.sota2.com/research/sota/text-to-audio-on-ritta-test.md?metric=mAMSR>): Higher is better; unit: SCALAR.

## Recorded results

| Method | FAD | Result date | Report paper |
| --- | --- | --- | --- |
| AudioLDM (L-Full) | 5.47 | Not recorded | 68fa767d1e209131cea6d5bc |
| AudioLDM (S-Full) | 5.65 | Not recorded | 68fa767d1e209131cea6d5bc |
| AudioGen | 6.43 | Not recorded | 68fa767d1e209131cea6d5bc |
| AudioLDM 2 (L-Full) | 6.68 | Not recorded | 68fa767d1e209131cea6d5bc |
| TangoFlux | 8.07 | Not recorded | 68fa767d1e209131cea6d5bc |
| TangoFlux | 8.93 | Not recorded | 68fa767d1e209131cea6d5bc |
| MakeAnAudio | 9.46 | Not recorded | 68fa767d1e209131cea6d5bc |
| Tango | 10.79 | Not recorded | 68fa767d1e209131cea6d5bc |
| Tango | 11.88 | Not recorded | 68fa767d1e209131cea6d5bc |
| Tango 2 | 13.84 | Not recorded | 68fa767d1e209131cea6d5bc |
| Tango 2 | 14.76 | Not recorded | 68fa767d1e209131cea6d5bc |

### AudioLDM (L-Full)

FAD: 5.47. Result ID: 69d89c5547f34ce924c0dd06.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70911e209131cea4be25.md>)

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

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"739 M"}. This record does not assert independent reproduction.

### AudioLDM (S-Full)

FAD: 5.65. Result ID: 69d89c5547f34ce924c0dd05.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70911e209131cea4be25.md>)

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

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"185 M"}. This record does not assert independent reproduction.

### AudioGen

FAD: 6.43. Result ID: 69d89c5547f34ce924c0dd09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb71e209131cea41b9a.md>)

[Implementation](<https://felixkreuk.github.io/audiogen>)

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"1.5 B"}. This record does not assert independent reproduction.

### AudioLDM 2 (L-Full)

FAD: 6.68. Result ID: 69d89c5547f34ce924c0dd07.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

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

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

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"844 M"}. This record does not assert independent reproduction.

### TangoFlux

FAD: 8.07. Result ID: 69d89c5547f34ce924c0dd0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

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

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

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"515 M"}. This record does not assert independent reproduction.

### TangoFlux

FAD: 8.93. Result ID: 69d89c5547f34ce924c0dd0f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

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

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

Recorded evaluation configuration: {"Methodology":"LLM+Agentic+TTA Model","#param":"515 M","LLM":"GPT-4"}. This record does not assert independent reproduction.

### MakeAnAudio

FAD: 9.46. Result ID: 69d89c5547f34ce924c0dd08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e711e209131cea3fc92.md>)

[Implementation](<https://text-to-audio.github.io/>)

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"452 M"}. This record does not assert independent reproduction.

### Tango

FAD: 10.79. Result ID: 69d89c5547f34ce924c0dd0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f771e209131cea45c2c.md>)

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

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"866 M"}. This record does not assert independent reproduction.

### Tango

FAD: 11.88. Result ID: 69d89c5547f34ce924c0dd0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f771e209131cea45c2c.md>)

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

Recorded evaluation configuration: {"Methodology":"LLM+Agentic+TTA Model","#param":"866 M","LLM":"GPT-4"}. This record does not assert independent reproduction.

### Tango 2

FAD: 13.84. Result ID: 69d89c5547f34ce924c0dd0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74211e209131cea5fd90.md>)

Recorded evaluation configuration: {"Methodology":"Zero-Shot","#param":"866 M"}. This record does not assert independent reproduction.

### Tango 2

FAD: 14.76. Result ID: 69d89c5547f34ce924c0dd0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa767d1e209131cea6d5bc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74211e209131cea5fd90.md>)

Recorded evaluation configuration: {"Methodology":"LLM+Agentic+TTA Model","#param":"866 M","LLM":"GPT-4"}. This record does not assert independent reproduction.

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