OpenCodePapers

speech-synthesis-on-libritts

Accented Speech RecognitionSpeech Synthesis
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodePESQM-STFTMCDPeriodicityV/UV F1ModelNameReleaseDate
Accelerating High-Fidelity Waveform Generation via Adversarial Flow Matching Optimization✓ Link4.4540.73580.05280.9756PeriodWave-Turbo-L2024-08-15
BigVGAN: A Universal Neural Vocoder with Large-Scale Training✓ Link4.3620.70260.29030.05930.9793BigVGAN-v22022-06-09
EVA-GAN: Enhanced Various Audio Generation via Scalable Generative Adversarial Networks✓ Link4.35360.79820.07510.9745EVA-GAN-big2024-01-31
PeriodWave: Multi-Period Flow Matching for High-Fidelity Waveform Generation✓ Link4.2481.02690.07650.9651PeriodWave + FreeU2024-08-14
RFWave: Multi-band Rectified Flow for Audio Waveform Reconstruction✓ Link4.2280.0900.968RFWave2024-03-08
BigVSAN: Enhancing GAN-based Neural Vocoders with Slicing Adversarial Network✓ Link4.1200.79920.41290.09240.9644BigVSAN (w/ snakebeta)2023-09-06
BigVSAN: Enhancing GAN-based Neural Vocoders with Slicing Adversarial Network✓ Link4.1160.78810.33810.09350.9635BigVSAN2023-09-06
EVA-GAN: Enhanced Various Audio Generation via Scalable Generative Adversarial Networks✓ Link4.03300.94850.09420.9658EVA-GAN-base2024-01-31
BigVGAN: A Universal Neural Vocoder with Large-Scale Training✓ Link4.0270.79970.37450.10180.9598BigVGAN2022-06-09
Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis✓ Link3.700.1010.9582Vocos2023-06-01
BigVGAN: A Universal Neural Vocoder with Large-Scale Training✓ Link3.5190.87880.45640.12870.9459BigVGAN-base2022-06-09
WaveGlow: A Flow-based Generative Network for Speech Synthesis✓ Link3.1381.30992.35910.14850.9378WaveGlow2018-10-31
WaveFlow: A Compact Flow-based Model for Raw Audio✓ Link3.0271.11201.24550.14160.9410WaveFlow2019-12-03
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis✓ Link2.9471.00170.6603 0.15650.9300HiFi-GAN2020-10-12
Speaker Conditional WaveRNN: Towards Universal Neural Vocoder for Unseen Speaker and Recording Conditions✓ Link1.7012.23581.88540.30440.8144SC-WaveRNN2020-08-09