Nonparallel training for voice conversion based on a parameter adaptation approach

The objective of voice conversion algorithms is to modify the speech by a particular source speaker so that it sounds as if spoken by a different target speaker. Current conversion algorithms employ a training procedure, during which the same utterances spoken by both the source and target speakers are needed for deriving the desired conversion parameters. Such a (parallel) corpus, is often difficult or impossible to collect. Here, we propose an algorithm that relaxes this constraint, i.e., the training corpus does not necessarily contain the same utterances from both speakers. The proposed algorithm is based on speaker adaptation techniques, adapting the conversion parameters derived for a particular pair of speakers to a different pair, for which only a nonparallel corpus is available. We show that adaptation reduces the error obtained when simply applying the conversion parameters of one pair of speakers to another by a factor that can reach 30%. A speaker identification measure is also employed that more insightfully portrays the importance of adaptation, while listening tests confirm the success of our method. Both the objective and subjective tests employed, demonstrate that the proposed algorithm achieves comparable results with the ideal case when a parallel corpus is available.     

Smaller,Stronger, More Stable: Peptide Variants of a SARS-CoV-2 Neutralizing Miniprotein

Based on the structure of a de novo designed miniprotein (LCB1) in complex with the receptor binding domain (RBD) of the SARS-CoV-2 spike protein, we have generated and characterized truncated peptide variants of LCB1, which present only two of the three LCB1 helices, and which fully retained the virus neutralizing potency against different SARS-CoV-2 variants of concern (VOC). This antiviral activity was even 10-fold stronger for a cyclic variant of the two-helix peptides, as compared to the full-length peptide. Furthermore, the proteolytic stability of the cyclic peptide was substantially improved, rendering it a better potential candidate for SARS-CoV-2 therapy. In a more mechanistic approach, the peptides also served as tools to dissect the role of individual mutations in the RBD for the susceptibility of the resulting virus variants to neutralization by the peptides. As the peptides reported here were generated through chemical synthesis, rather than recombinant protein expression, they are amenable to further chemical modification, including the incorporation of a wide range of non-proteinogenic amino acids, with the aim to further stabilize the peptides against proteolytic degradation, as well as to improve the strength, as well the breadth, of their virus neutralizing capacity.     

Relationship between microscopic and macroscopic structures of organic thin films for SHG

The experiments described in this paper were undertaken in order to obtain information about the relationship between the structure and non-linear optical properties (second-harmonic generation) of organic thin films. For this purpose, two closely related dyes, diones and tetrones, were compared, both of which are shown to have large hyperpolarisabilities. Their microscopic properties are investigated by conformational analysis and electron diffraction. It could be shown that detailed knowledge about the structure and the adjacent neighbour packing can be obtained from conformational analysis and electron diffraction in order to understand the non-linear optical properties of the two dyes.     

Cloning and Functional Characterization of Dog OCT1 and OCT2: Another Step in Exploring Species Differences in Organic Cation Transporters

OCT1 and OCT2 are polyspecific membrane transporters that are involved in hepatic and renal drug clearance in humans and mice. In this study, we cloned dog OCT1 and OCT2 and compared their function to the human and mouse orthologs. We used liver and kidney RNA to clone dog OCT1 and OCT2. The cloned and the publicly available RNA-Seq sequences differed from the annotated exon-intron structure of OCT1 in the dog genome CanFam3.1. An additional exon between exons 2 and 3 was identified and confirmed by sequencing in six additional dog breeds. Next, dog OCT1 and OCT2 were stably overexpressed in HEK293 cells and the transport kinetics of five drugs were analyzed. We observed strong differences in the transport kinetics between dog and human orthologs. Dog OCT1 transported fenoterol with 12.9-fold higher capacity but 14.3-fold lower affinity (higher K_M) than human OCT1. Human OCT1 transported ipratropium with 5.2-fold higher capacity but 8.4-fold lower affinity than dog OCT1. Compared to human OCT2, dog OCT2 showed 10-fold lower transport of fenoterol and butylscopolamine. In conclusion, the functional characterization of dog OCT1 and OCT2 reported here may have implications when using dogs as pre-clinical models as well as for drug therapy in dogs.     

Decarbonylation Kinetics of Formic Acid in Aqueous Sulphuric Acid Solutions

Decarbonylation of formic acid in 46.6–80.2% (6.5–14.1 m) sulphuric acid was studied by driving CO out of the reaction medium by an air current followed by its photometrical determination with Ag-Na-sulphamoyl-benzoate. The logarithm of the reaction rate constant increases straight with the acidity of the medium in the 6.5–12.6 M H₂SO₄ range. The data obtained argue in favour of a possible implication of a sulphuric acid molecule respectively hydrogensulphate ion in a rate controlling step.     

How to learn from the resilience of Human–Machine Systems?

This paper proposes a functional architecture to learn from resilience. First, it defines the concept of resilience applied to Human–Machine System (HMS) in terms of safety management for perturbations and proposes some indicators to assess this resilience. Local and global indicators for evaluating human–machine resilience are used for several criteria. A multi-criteria resilience approach is then developed in order to monitor the evolution of local and global resilience. The resilience indicators are the possible inputs of a learning system that is capable of producing several outputs, such as predictions of the possible evolutions of the system's resilience and possible alternatives for human operators to control resilience. Our system has a feedback–feedforward architecture and is capable of learning from the resilience indicators. A practical example is explained in detail to illustrate the feasibility of such prediction.     

Mini-maximizing two qubit quantum computations

Two qubit quantum computations are viewed as two player, strictly competitive games and a game-theoretic measure of optimality of these computations is developed. To this end, the geometry of Hilbert space of quantum computations is used to establish the equivalence of game-theoretic solution concepts of Nash equilibrium and mini-max outcomes in games of this type, and quantum mechanisms are designed for realizing these mini-max outcomes.