We propose a parallel-data-free voice conversion (VC)method that can learn a mapping from source to target speech without relying on parallel data. The proposed method is generalpurpose, high quality, and parallel-data-free, which works without any extra data, modules, or alignment procedure. It is also noteworthy that it avoids over-smoothing, which occurs in many conventional statistical mode...

It is traditionally assumed that various sources of linguistic knowledge and their interaction should be formalised in order to be able to convert words into their phonemic representations with reasonable accuracy. We show that using supervised learning techniques, based on a corpus of transcribed words, the same and even better performance can be achieved, without explicit modeling of linguist...

The POOL data storage mechanism is intended to satisfy the needs of the LHC experiments to store and analyze the data from the detector response of particle collisions at the LHC proton-proton collider. Both the data rate and the data volumes will largely differ from the past experience. The POOL data storage mechanism is intended to be able to cope with the experiment’s requirements applying a...

A two-level data transformation consists of a type-level transformation of a data format coupled with value-level transformations of data instances corresponding to that format. We have implemented a system for performing two-level transformations on XML schemas and their corresponding documents, and on SQL schemas and the databases that they describe. The core of the system consists of a combi...

UNLABELLED fconv is a program intended for parsing and manipulating multiple aspects and properties of molecular data. Up to now, it has been developed and extensively tested for 3 years. It has become a very robust and comprehensive tool involved in a broad range of computational workflows that are currently applied in our drug design environment. Typical tasks are as follows: conversion and e...

In the traditional (knowledge-based) approach to the design of grapheme-to-phoneme modules in text-to-speech systems, it is claimed that various explicitly coded, language-speciic, linguistic knowledge sources are necessary for a good performance. Due to knowledge acquisition bottlenecks, this implies long development cycles. As an alternative, we propose to use inductive methods from machine l...

Transparent optical multicast by multi-wavelength conversion has revealed a brand-new way for performing data multicast function directly in the optical domain without passing through any electronics. In this paper, several simultaneous multiple wavelength conversion approaches for transparent optical multicast are discussed. Technologies and methods for single-to-multiple-channel wavelength co...