Text-to-sound-effects generation on Adobe Audition SFX 250 hand-written text prompts
70.67Text-audio similarityDAC-VAE
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DAC-VAEKL=200.39, Bitrate=11.562026.05 | 70.67 | 1.7 | 0.11 | |
| DAC-VAEKL=132.63, Bitrate=7.652026.05 | 69.76 | 1.93 | 0.11 | |
| DAC-VAEKL=642.35, Bitrate=37.062026.05 | 68.99 | 2.02 | 0.12 | |
| DAC-VAEKL=341.26, Bitrate=19.692026.05 | 68.8 | 2.28 | 0.12 | |
| Stable Audio Open (SAO)KL=82.16, Bitrate=4.742026.05 | 68.38 | 2.13 | 0.13 | |
| DAC-VAEKL=1284.21, Bitrate=74.102026.05 | 66.84 | 2.16 | 0.12 |