Audio Generation On Audiocaps

평가 지표

FAD

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	FAD	FD	Paper Title	Repository
Make-An-Audio 2	1.80	11.75	Make-An-Audio 2: Temporal-Enhanced Text-to-Audio Generation
Stable Audio	-	-	Fast Timing-Conditioned Latent Audio Diffusion
Audiobox Sound	0.77	8.30	Audiobox: Unified Audio Generation with Natural Language Prompts	-
GenAu-Large	1.21	16.51	Taming Data and Transformers for Audio Generation
Tango-AF&AC-FT-AC	2.54	17.19	Improving Text-To-Audio Models with Synthetic Captions
AudioLDM 2-AC-Large	1.42	-	AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
Re-AudioLDM-L	1.37	-	Retrieval-Augmented Text-to-Audio Generation	-
Auffusion-Full	1.76	23.08	Auffusion: Leveraging the Power of Diffusion and Large Language Models for Text-to-Audio Generation
ETTA	2.51	13.12	ETTA: Elucidating the Design Space of Text-to-Audio Models	-
AudioGen	3.13	-	AudioGen: Textually Guided Audio Generation
Make-An-Audio	2.66	18.32	Make-An-Audio: Text-To-Audio Generation with Prompt-Enhanced Diffusion Models
TangoFlux	-	-	TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization
ETTA-FT-AC-100k	2.03	10.10	ETTA: Elucidating the Design Space of Text-to-Audio Models	-
Diffsound	7.75	47.68	Diffsound: Discrete Diffusion Model for Text-to-sound Generation
AudioLDM2-large	2.02	26.18	AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
Stable Audio 2.0	-	-	Long-form music generation with latent diffusion
Consistency TTA (Single-step generation)	2.18	20.44	ConsistencyTTA: Accelerating Diffusion-Based Text-to-Audio Generation with Consistency Distillation	-
Auffusion	1.63	21.99	Auffusion: Leveraging the Power of Diffusion and Large Language Models for Text-to-Audio Generation
CoDi	1.80	22.90	Any-to-Any Generation via Composable Diffusion
AudioLDM-L-Full	1.96	23.31	AudioLDM: Text-to-Audio Generation with Latent Diffusion Models

0 of 23 row(s) selected.