Research map / Image generation / GANs
Adversarial Generation of Time-Frequency Features with application in audio synthesis
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Generate log-magnitude STFT features with TiFGAN-M rather than waveform samples with WaveGAN | TiFGAN-M vs WaveGAN · Speech Commands sc09 | preference: 75% | Helped |
| Generate log-magnitude STFT features with TiFGAN-M rather than waveform samples with WaveGAN | TiFGAN-M vs WaveGAN · Speech Commands sc09 | inception score: 5.97 vs 4.64 | Helped |
| Generate log-magnitude STFT features with TiFGAN-M rather than waveform samples with WaveGAN | TiFGAN-M vs WaveGAN · Speech Commands sc09 | FID: 41.6 -> 26.7 | Helped |
| Add generated time- and frequency-direction phase derivatives (ablation) | TiFGAN-M · Speech Commands sc09 | inception score: 5.97 -> 4.48 | Hurt |
| Add generated time- and frequency-direction phase derivatives (ablation) | TiFGAN-M · Speech Commands sc09 | FID: 26.7 -> 32.6 | Hurt |
| Add generated time- and frequency-direction phase derivatives (ablation) | TiFGAN-M · Speech Commands sc09 | preference over WaveGAN: 75% vs 55% | Hurt |
| Reconstruct phase from magnitude using PGHI instead of generating phase derivatives (ablation) | TiFGAN-MTF phase reconstruction · Speech Commands sc09 | RSPE (dB): -7.5 dB vs -12.5* | Helped |
Closest papers by meaning
- GANSynth: Adversarial Neural Audio Synthesis (ICLR 2019 · similarity 0.90)
- Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (ICLR 2024 · similarity 0.90)
- Adversarial Audio Synthesis (ICLR 2019 · similarity 0.89)
- Fast Timing-Conditioned Latent Audio Diffusion (ICML 2024 · similarity 0.86)
- PeriodWave: Multi-Period Flow Matching for High-Fidelity Waveform Generation (ICLR 2025 · similarity 0.85)
- Representing Sounds as Neural Amplitude Fields: A Benchmark of Coordinate-MLPs and a Fourier Kolmogorov-Arnold Framework (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.