AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Toward Complex-Valued Neural Networks for Waveform Generation

Hyung-Seok Oh, Deok-Hyeon Cho, Seung-Bin Kim, Seong-Whan Lee

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Diffusion audio synthesis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
use complex-valued networks instead of real-valued (ablation)MLP GAN · Synthetic complex distributionJSD (magnitude): 0.018350 -> 0.006548Improved
use complex-valued networks instead of real-valued (ablation)MLP GAN · Synthetic complex distributionJSD (phase): 0.021110 -> 0.003911Improved
make generator complex-valued (ablation)ComVo · LibriTTSUTMOS: 3.6025 -> 3.6452Improved
make discriminator complex-valued (cMRD) with real generator (ablation)Vocos-like · LibriTTSUTMOS: 3.6025 -> 3.5930No clear effect
make both generator and discriminator complex-valued (ablation)Vocos-like · LibriTTSUTMOS: 3.6025 -> 3.6646Improved
add phase quantization Nq=128 (ablation)ComVo · LibriTTSUTMOS: 3.6646 -> 3.6901Improved
add phase quantization Nq=128 (ablation)ComVo · LibriTTSMR-STFT: 0.8435 -> 0.8439Hurt
increase quantization levels from 128 to 512 (ablation)ComVo · LibriTTSUTMOS: 3.6901 -> 3.6412Hurt
use block-matrix computation instead of native PyTorch complex ops (ablation)ComVo · LibriTTStraining time (hrs): 183 -> 138Improved
use block-matrix computation instead of native PyTorch complex ops (ablation)ComVo · LibriTTSGPU xRT: 702.26 -> 696.91Hurt
double real-valued model parameters instead of complex modeling (ablation)G_R D_R · LibriTTSUTMOS: 3.6025 -> 3.6164No clear effect
add phase quantization after first complex conv (ablation)ComVo · LibriTTSUTMOS: 3.6646 -> 3.6901Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy