Diffusion models have recently become a popular generative modeling framework in various domains because of their high-quality sampling capabilities. Lately, it has been hypothesized that optimally trained diffusion models supplied with specific differential equation solvers provide a solution to the optimal transport problem between the data distribution and the prior distribution. In this paper, we empirically show that applying the optimal transport point of view on diffusion modeling allows making a good choice of a noise sample the reverse diffusion starts generating from. We consider two audio-related tasks: voice conversion and timbre transfer. In the former, we improve upon the recent state-of-the-art model and demonstrate that the optimal transport helps us to keep the prosody of the source utterances significantly better than the vanilla diffusion-based model does. As for timbre transfer, we propose the novel diffusion model capable of many-to-many timbre transfer performing on par with common algorithms in terms of the overall music quality.

SPHINCS+ [CCS ’19] is one of the selected post-quantum digital signature schemes of NIST’s post-quantum standardization process. The scheme is a hash-based signature and is considered one of the most secure and robust proposals. The proposal includes a fast (but larger) variant and a small (but slower) variant for each security level. The main problem that might hinder its adoption is its large signature size. Although SPHINCS+ supports a trade-off between signature size and the computational cost of signing, further reducing the signature size (below the small variants) results in a prohibitively high computational cost for the signer.This paper presents several novel methods for further compressing the signature size while requiring negligible added computational costs for the signer and further reducing verification time. Moreover, our approach enables a much more efficient trade-off curve between signature size and the computational costs of the signer. In many parameter settings, we achieve small signatures and faster running times simultaneously. For example, for 128-bit (classical) security, the small signature variant of SPHINCS+ is 7856 bytes long, while our variant is only 6304 bytes long: a compression of approximately 20% while still reducing the signer’s running time. However, other trade-offs that focus, e.g., on verification speed, are possible.The main insight behind our scheme is that there are predefined specific subsets of messages for which the WOTS+ and FORS signatures (that SPHINCS+ uses) can be compressed, and generation can be made faster while maintaining the same security guarantees. Although most messages will not come from these subsets, we can search for suitable hashed values to sign. We sign a hash of the message concatenated with a counter that was chosen such that the hashed value is in the subset. The resulting signature is both smaller and faster to sign and verify.Our schemes are simple to describe and implement. We provide an implementation, a theoretical analysis of speed and security, as well as benchmark results.

Recently, denoising diffusion probabilistic models and generative score matching have shown high potential in modelling complex data distributions while stochastic calculus has provided a unified point of view on these techniques allowing for flexible inference schemes. In this paper we introduce Grad-TTS, a novel text-to-speech model with score-based decoder producing mel-spectrograms by gradually transforming noise predicted by encoder and aligned with text input by means of Monotonic Alignment Search. The framework of stochastic differential equations helps us to generalize conventional diffusion probabilistic models to the case of reconstructing data from noise with different parameters and allows to make this reconstruction flexible by explicitly controlling trade-off between sound quality and inference speed. Subjective human evaluation shows that Grad-TTS is competitive with state-of-the-art text-to-speech approaches in terms of Mean Opinion Score.

In the present work, a peculiar property of hash-based signatures allowing detection of their forgery event is explored. This property relies on the fact that a successful forgery of a hash-based signature most likely results in a collision with respect to the employed hash function, while the demonstration of this collision could serve as convincing evidence of the forgery. Here we prove that with properly adjusted parameters Lamport and Winternitz one-time signatures schemes could exhibit a forgery detection availability property. This property is of significant importance in the framework of crypto-agility paradigm since the considered forgery detection serves as an alarm that the employed cryptographic hash function becomes insecure to use and the corresponding scheme has to be replaced.

LPCNet vocoder has recently been presented to TTS community and is now gaining increasing popularity due to its effectiveness and high quality of the speech synthesized with it. In this work, we present a modification of LPCNet that is 1.5x faster, has twice less non-zero parameters and synthesizes speech of the same quality. Such enhancement is possible mostly due to two features that we introduce into the original architecture: the proposed vocoder is designed to generate 16-bit signal instead of 8-bit μ-companded signal, and it predicts two consecutive excitation values at a time independently of each other. To show that these modifications do not lead to quality degradation we train models for five different languages and perform extensive human evaluation.