One hand-crafted topological feature makes speech neural nets far more noise-robust
bravo_abad · x · 2026-09-21
A new paper by Feng et al. shows that converting a speech signal into a geometric object via time-delay embedding and using persistent homology to measure the persistence of its dominant structures yields a single hand-crafted topological feature that markedly improves a speech network's noise robustness — challenging the assumption that deep nets always learn such features themselves. The author also runs the Substack newsletter Discovery at Scale, curating actionable AI-for-Science ideas weekly.
Related event: Handcrafted Topological Feature Boosts Speech Network Noise Robustness(2 posts)→
More from Research
- Flow Matching revisited: the simulation-free CNF training method behind today's generative models — burny_tech · 2026-09-21
- DIY-Jev: no fine-tuning needed — reading true/false logits hits 75.5% on 32k benchmark — Malfeitor1235 · 2026-09-21
- Quadruped locomotion policy adds heat management to cut motor overheating in real deployments — IsaiahBallah · 2026-09-21
- LIVIA Launches as Browser Tool to Score Protein-Protein Interface Predictions — sokrypton · 2026-09-21
- Cybernetics vs. Minsky's AI: deep learning was reborn in the automata universe, says Plinz — burny_tech · 2026-09-21
- First RNA therapy success: man with rare ALS improves after mutation-targeting treatment — burny_tech · 2026-09-21