Text-to-Speech on LJSpeech (test)

86.67CMOS

OscillaTTS

Updated 1mo ago

Evaluation Results

Method
OscillaTTS 2026.06	86.67	-	-	-	-	6.59	0.35
GRADTTS 2026.06	83.78	-	-	-	-	6.9	0.35
StyleTTS2 2026.06	81.48	-	-	-	-	6.64	0.41
FASTSPEECH2 2026.06	76	-	-	-	-	6.62	0.35
GlowTTS 2026.06	75.79	-	-	-	-	6.85	0.4
Glow-TTS 2020.05	0.934	-	-	-	-	-	-
Tacotron 2 2020.05	0	-	-	-	-	-	-
NaturalSpeech 2022.05	0	-	-	-	-	-	-
Human Recordings 2022.05	0	-	-	-	-	-	-
Human Recordings 2022.05	0	-	-	-	-	-	-
FastSpeech 2 2020.06	0	-	-	-	-	-	-
FastSpeech 2 2020.06	0	3.83	-	-	-	-	-
NaturalSpeech 2022.05	-0.01	-	0.6902	-	-	-	-
VITS 2022.05	-0.19	-	0.2429	-	-	-	-
VITS 2022.05	-0.2	-	-	-	-	-	-
Glow-TTS + HiFiGAN 2022.05	-0.23	-	0	-	-	-	-
Grad-TTS + HiFiGAN 2022.05	-0.23	-	0.0127	-	-	-	-
Transformer TTS 2020.06	-0.235	-	-	-	-	-	-
Transformer TTS 2020.06	-0.235	3.72	-	-	-	-	-
Grad-TTS 2022.05	-0.24	-	-	-	-	-	-
Glow-TTS 2022.05	-0.26	-	-	-	-	-	-
FastSpeech 2 + HiFiGAN 2022.05	-0.3	-	0	-	-	-	-
FastSpeech 2 2022.05	-0.33	-	-	-	-	-	-
FastSpeech 2020.06	-0.885	-	-	-	-	-	-
FastSpeech 2020.06	-0.885	3.68	-	-	-	-	-
GT 2020.06	-	4.3	-	-	-	-	-
GT 2020.06	-	3.92	-	-	-	-	-
Tacotron 2 2020.06	-	3.7	-	-	-	-	-
FastSpeech 2s 2020.06	-	3.71	-	-	-	-	-
T2 2022.11	-	3.25	-	-	-	-	-
GTTS 2022.11	-	2.64	-	-	-	-	-
NHMM 2022.11	-	2.97	-	-	-	-	-
OverFlow (OF) 2022.11	-	3.43	-	-	-	-	-
OverFlow (OFND) 2022.11	-	3.25	-	-	-	-	-
OverFlow (OFZT) 2022.11	-	3.01	-	-	-	-	-
VOC 2022.11	-	4.18	-	-	-	-	-
Ground Truth (LJSpeech) 2026.04	-	4.51	-	0.08	-	-	-
Baseline (FS2 + PWG) 2026.04	-	2.81	-	0.13	-	-	-
Proposed Method (FS2 + Mimi) 2026.04	-	2.51	-	0.11	-	-	-
Topline (Mimi Reconstruction) 2026.04	-	-	-	-	1.34	-	-
VITS (Baseline) 2026.04	-	-	-	-	1.64	-	-
FastSpeech 2 (Baseline) 2026.04	-	-	-	-	5.19	-	-
Proposed Method (FS2 + Mimi) 2026.04	-	-	-	-	8.89	-	-