SOTA Audio Generation benchmarks and papers with code

Benchmarks

Dataset Name	SOTA Method	Metric
LJ Speech (test)	Autoregressive flow	LL Score5.161	20	4d ago
Speech clean (test)		Generation Success Rate100	18	4d ago
Speech Noise SNR=-10 (test)	SEEN	Success Rate80	18	4d ago
Sound Clean (test)		Generation Success Rate100	18	4d ago
Sound Noise SNR=-10 (test)	SEEN	Success Rate78.33	18	4d ago
Music clean (test)		Generation Success Rate100	18	4d ago
Music Noise SNR=-10 (test)	SEEN	Generation Success Rate85	18	4d ago
LibriTTS (dev)	HiFi-GAN	M-STFT1.3647	18	4d ago
MMAU Speech (Noise)	SEEN	GSR80	15	4d ago
MMAU Speech (Clean)	SEEN	GSR99	15	4d ago
MMAU Sound (Noise)	SEEN	GSR78.33	15	4d ago
MMAU Sound (Clean)	SEEN	GSR99.67	15	4d ago
MMAU Music (Noise)	SEEN	GSR85	15	4d ago
MMAU Music (Clean)	SEEN	GSR99	15	4d ago
EpicBench T2A 1.0 (test)	T2A-Feedback + DPO	Win Rate EOS68	8	4d ago
HDTF	Narrating For You	FAD106.43	5	4d ago
CelebV-HQ	Narrating For You	FAD244.83	5	4d ago
FakeAVCeleb	Narrating For You	FAD171.52	5	4d ago
VoxCeleb	Your_TTS	FAD199.52	5	4d ago
Audio generation (evaluation set)	AudioGen	KL Divergence1.42	4	4d ago
JavisBench mini OoD 4s (test)	JavisDiT-audio-ep55	FAD8.11	3	4d ago
AudioCaps InD, 10s (test)	AudioLDM2	FAD2.01	3	4d ago
TTS 80-ms audio segments (test)	T-Mimi	Latency (ms)4.4	3	4d ago
AudioCaps InD, 4s (test)	AudioLDM2	FAD5.57	2	4d ago
Piano (test)	Wavenet	Bits Per Character1.88	2	4d ago

Showing 25 of 26 rows