Yang Li

Siyuan Shan, Yang Li, Amartya Banerjee et al.

Voice conversion (VC) aims at altering a person's voice to make it sound similar to the voice of another person while preserving linguistic content. Existing methods suffer from a dilemma between content intelligibility and speaker similarity; i.e., methods with higher intelligibility usually have a lower speaker similarity, while methods with higher speaker similarity usually require plenty of target speaker voice data to achieve high intelligibility. In this work, we propose a novel method \textit{Phoneme Hallucinator} that achieves the best of both worlds. Phoneme Hallucinator is a one-shot VC model; it adopts a novel model to hallucinate diversified and high-fidelity target speaker phonemes based just on a short target speaker voice (e.g. 3 seconds). The hallucinated phonemes are then exploited to perform neighbor-based voice conversion. Our model is a text-free, any-to-any VC model that requires no text annotations and supports conversion to any unseen speaker. Objective and subjective evaluations show that \textit{Phoneme Hallucinator} outperforms existing VC methods for both intelligibility and speaker similarity.

Yang Li, Hongbo Li

Solving random subset sum instances plays an important role in constructing cryptographic systems. For the random subset sum problem, in 2013 Bernstein et al. proposed a quantum algorithm with heuristic time complexity $\widetilde{O}(2^{0.241n})$, where the "$\widetilde{O}$" symbol is used to omit poly($\log n$) factors. In 2018, Helm and May proposed another quantum algorithm that reduces the heuristic time and memory complexity to $\widetilde{O}(2^{0.226n})$. In this paper, a new quantum algorithm is proposed, with heuristic time and memory complexity $\widetilde{O}(2^{0.209n})$.