Showing 1–3 of 3 results for author: Pawlicki, A

CCATMos: Convolutional Context-aware Transformer Network for Non-intrusive Speech Quality Assessment

Authors: Yuchen Liu, Li-Chia Yang, Alex Pawlicki, Marko Stamenovic

Abstract: Speech quality assessment has been a critical component in many voice communication related applications such as telephony and online conferencing. Traditional intrusive speech quality assessment requires the clean reference of the degraded utterance to provide an accurate quality measurement. This requirement limits the usability of these methods in real-world scenarios. On the other hand, non-in… ▽ More Speech quality assessment has been a critical component in many voice communication related applications such as telephony and online conferencing. Traditional intrusive speech quality assessment requires the clean reference of the degraded utterance to provide an accurate quality measurement. This requirement limits the usability of these methods in real-world scenarios. On the other hand, non-intrusive subjective measurement is the ``golden standard" in evaluating speech quality as human listeners can intrinsically evaluate the quality of any degraded speech with ease. In this paper, we propose a novel end-to-end model structure called Convolutional Context-Aware Transformer (CCAT) network to predict the mean opinion score (MOS) of human raters. We evaluate our model on three MOS-annotated datasets spanning multiple languages and distortion types and submit our results to the ConferencingSpeech 2022 Challenge. Our experiments show that CCAT provides promising MOS predictions compared to current state-of-art non-intrusive speech assessment models with average Pearson correlation coefficient (PCC) increasing from 0.530 to 0.697 and average RMSE decreasing from 0.768 to 0.570 compared to the baseline model on the challenge evaluation test set. △ Less

Submitted 4 November, 2022; originally announced November 2022.

Weight, Block or Unit? Exploring Sparsity Tradeoffs for Speech Enhancement on Tiny Neural Accelerators

Authors: Marko Stamenovic, Nils L. Westhausen, Li-Chia Yang, Carl Jensen, Alex Pawlicki

Abstract: We explore network sparsification strategies with the aim of compressing neural speech enhancement (SE) down to an optimal configuration for a new generation of low power microcontroller based neural accelerators (microNPU's). We examine three unique sparsity structures: weight pruning, block pruning and unit pruning; and discuss their benefits and drawbacks when applied to SE. We focus on the int… ▽ More We explore network sparsification strategies with the aim of compressing neural speech enhancement (SE) down to an optimal configuration for a new generation of low power microcontroller based neural accelerators (microNPU's). We examine three unique sparsity structures: weight pruning, block pruning and unit pruning; and discuss their benefits and drawbacks when applied to SE. We focus on the interplay between computational throughput, memory footprint and model quality. Our method supports all three structures above and jointly learns integer quantized weights along with sparsity. Additionally, we demonstrate offline magnitude based pruning of integer quantized models as a performance baseline. Although efficient speech enhancement is an active area of research, our work is the first to apply block pruning to SE and the first to address SE model compression in the context of microNPU's. Using weight pruning, we show that we are able to compress an already compact model's memory footprint by a factor of 42x from 3.7MB to 87kB while only losing 0.1 dB SDR in performance. We also show a computational speedup of 6.7x with a corresponding SDR drop of only 0.59 dB SDR using block pruning. △ Less

Submitted 9 November, 2021; v1 submitted 3 November, 2021; originally announced November 2021.

Comments: To appear in NeurIPS 2021 Efficient Natural Langauge and Speech Processing Workshop as oral-spotlight presentation

LaAlO3 stoichiometry found key to electron liquid formation at LaAlO3/SrTiO3 interfaces

Authors: M. P. Warusawithana, C. Richter, J. A. Mundy, P. Roy, J. Ludwig, S. Paetel, T. Heeg, A. A. Pawlicki, L. F. Kourkoutis, M. Zheng, M. Lee, B. Mulcahy, W. Zander, Y. Zhu, J. Schubert, J. N. Eckstein, D. A. Muller, C. Stephen Hellberg, J. Mannhart, D. G. Schlom

Abstract: Emergent phenomena, including superconductivity and magnetism, found in the two-dimensional electron liquid (2-DEL) at the interface between the insulators LaAlO3 and SrTiO3 distinguish this rich system from conventional two-dimensional electron gases at compound semiconductor interfaces. The origin of this 2-DEL, however, is highly debated with focus on the role of defects in the SrTiO3 while the… ▽ More Emergent phenomena, including superconductivity and magnetism, found in the two-dimensional electron liquid (2-DEL) at the interface between the insulators LaAlO3 and SrTiO3 distinguish this rich system from conventional two-dimensional electron gases at compound semiconductor interfaces. The origin of this 2-DEL, however, is highly debated with focus on the role of defects in the SrTiO3 while the LaAlO3 has been assumed perfect. Our experiments and first principles calculations show that the cation stoichiometry of the nominal LaAlO3 layer is key to 2-DEL formation: only Al-rich LaAlO3 results in a 2-DEL. While extrinsic defects including oxygen deficiency are known to render LaAlO3/SrTiO3 samples conducting, our results show that in the absence of such extrinsic defects, an interface 2-DEL can form. Its origin is consistent with an intrinsic electronic reconstruction occurring to counteract a polarization catastrophe. This work provides a roadmap for identifying other interfaces where emergent behaviors await discovery. △ Less

Submitted 20 March, 2013; originally announced March 2013.