Text-to-Audio on RiTTA (test)
5.47FADAudioLDM (L-Full)
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| AudioLDM (L-Full)Methodology=Zero-Shot, #param=739 M2024.12 | 5.47 | 38.42 | 37.96 | 3.09 | 0.77 | 2.56 | 0.08 | |
| AudioLDM (S-Full)Methodology=Zero-Shot, #param=185 M2024.12 | 5.65 | 38.95 | 37.3 | 2.76 | 0.5 | 2.52 | 0.04 | |
| AudioGenMethodology=Zero-Shot, #param=1.5 B2024.12 | 6.43 | 28.01 | 32.04 | 9.61 | 2.12 | 8.6 | 2.27 | |
| AudioLDM 2 (L-Full)Methodology=Zero-Shot, #param=844 M2024.12 | 6.68 | 29.07 | 35.85 | 12.26 | 2.41 | 10.01 | 3.39 | |
| TangoFluxMethodology=Zero-Shot, #param=515 M2024.12 | 8.07 | 32.8 | 47.92 | 33.83 | 7.02 | 29.01 | 76.57 | |
| TangoFluxMethodology=LLM+Agentic+TTA Model, #param=515 M, LLM=GPT-42024.12 | 8.93 | 32.99 | 49 | 35.19 | 9.69 | 28.22 | 79.43 | |
| MakeAnAudioMethodology=Zero-Shot, #param=452 M2024.12 | 9.46 | 82.72 | 45.98 | 8.14 | 1.68 | 6.47 | 1.02 | |
| TangoMethodology=Zero-Shot, #param=866 M2024.12 | 10.79 | 90.26 | 39.46 | 11.13 | 2.27 | 9.88 | 3.1 | |
| TangoMethodology=LLM+Agentic+TTA Model, #param=866 M, LLM=GPT-42024.12 | 11.88 | 92.19 | 41.44 | 12.33 | 9.21 | 11.28 | 19.17 | |
| Tango 2Methodology=Zero-Shot, #param=866 M2024.12 | 13.84 | 89.66 | 44.03 | 16.63 | 4.4 | 12.53 | 11.55 | |
| Tango 2Methodology=LLM+Agentic+TTA Model, #param=866 M, LLM=GPT-42024.12 | 14.76 | 93.1 | 44.89 | 19.33 | 9.37 | 14.13 | 20.31 |