Text-to-Audio Generation

Benchmarks

Dataset Name	SOTA Method	Metric
AudioCaps (test)	AUDIOBOX SOUND	FAD0.77	154	4d ago
Clotho (test)	Make-An-Audio	FID17.23	17	1mo ago
AudioCaps (evaluation)	Re-AudioLDM	FAD1.85	11	24d ago
TTA-Bench Accuracy	TangoFlux	CE Score3.539	10	1mo ago
MusicCaps	Stable Audio	FDopenl3108.69	10	1mo ago
Customized-Fantasy	DreamAudio-DiT-L	PQ6.42	8	24d ago
Customized-Overlay	DreamAudio-L	FAD0.73	8	24d ago
Customized-Concatenation	DreamAudio-L	FAD0.46	8	24d ago
FSD50K		FD25.97	8	1mo ago
InternalSFX (test)	Woosh-Flow-Private	FD246.9	6	16d ago
evaluation benchmarks one-to-one	TangoFlux	CLAP Score42.71	6	1mo ago
One-to-one evaluation benchmarks Text-to-Audio	TangoFlux	FAD1.41	6	1mo ago
AudioCaption (test)		CLAP Score0.526	6	1mo ago
VGGSound	Siren + Plan-Critic	CLAP Score (Overall)35.88	5	1mo ago
RiTTA Count OOD (test)	TangoFlux-RAG	KL Divergence2.18	4	1mo ago
EpicBench T2A	Make an Audio 2	EOS19.96	4	1mo ago
Human Evaluation Subjective Audio Assessment (test)	TANGOFLUX	Z-Score (OVL)0.2486	4	1mo ago
MusicCaps	Stable Audio	Structure: Intro92.1	4	1mo ago
AudioSet (eval)	AudioLDM-L-Full	FD24.26	4	1mo ago
Kling-Audio Eval	Omni2Sound	KL Divergence2.36	3	1mo ago
AudioCaps 5s (test)	Make-An-Audio 2	FD (Fréchet Distance)13.78	3	1mo ago
Clotho eval 10s (test)	Make-An-Audio 2	FD19.97	3	1mo ago
AudioCaps	Stable Audio	Stereo Correctness0.57	1	1mo ago

Showing 23 of 23 rows