Search papers, labs, and topics across Lattice.
4
0
3
5
HALO reduces computational costs by 50% without sacrificing performance, revolutionizing lightweight speech enhancement models.
Achieve state-of-the-art universal speech enhancement with UniPASE, a model that minimizes linguistic hallucinations by converting degraded waveforms into clean phonetic representations.
Generative and predictive models can be fused to achieve state-of-the-art speech enhancement, outperforming single-branch approaches in robustness and perceptual quality.
Studio-quality speech enhancement without hallucination is now possible, thanks to a clever combination of dry-target finetuning and flow-matching.
