Per Hedelin

2 accepted papers

2025

Audio Decoding by Inverse Problem Solving

Pedro J. Villasana T., Lars F. Villemoes, Janusz Klejsa, Per Hedelin

ICASSP 2025accepted

We consider audio decoding as an inverse problem and solve it through diffusion posterior sampling. Explicit conditioning functions are developed for input signal measurements provided by an example of a transform domain perceptual audio codec. Viability is demonstrated by evaluating arbitrary pairi…

Cited by 0 Source ↗Scholar ↗

2019

High-quality Speech Coding with Sample RNN

Janusz Klejsa, Per Hedelin, Cong Zhou, Roy Fejgin, Lars F. Villemoes

ICASSP 2019accepted

We provide a speech coding scheme employing a generative model based on SampleRNN that, while operating at significantly lower bitrates, matches or surpasses the perceptual quality of state-of-the-art classic wide-band codecs. Moreover, it is demonstrated that the proposed scheme can provide a meani…

Cited by 0 Source ↗Scholar ↗