We propose a nonparallel data-driven emotional speech conversion method. It enables the transfer of emotion-related characteristics of a speech signal while preserving the speaker's identity and linguistic content. Most existing approaches require parallel data and time alignment, which is not available in many real applications. We achieve nonparallel training based on an unsupervised style transfer technique, which learns a translation model between two distributions instead of a deterministic one-to-one mapping between paired examples. The conversion model consists of an encoder and a decoder for each emotion domain. We assume that the speech signal can be decomposed into an emotion-invariant content code and an emotion-related style code in latent space. Emotion conversion is performed by extracting and recombining the content code of the source speech and the style code of the target emotion. We tested our method on a nonparallel corpora with four emotions. The evaluation results show the effectiveness of our approach.

Abstract:In this paper we provide an overview of decision frameworks aimed at crafting an energy technologyResearch & Development portfolio, based on the results of three large expert elicitation studies and a large scale energy-economic model. We introduce importance sampling as a technique for integrating elicitation data and large IAMs into decision making under uncertainty models. We show that it is important to include both parts of this equation -the prospects for technological advancement and the interactions of the technologies in and with the economy. We find that investment in energy technology R&D is important even in the absence of climate policy. We illustrate the value of considering dynamic two-stage sequential decision models under uncertainty for identifying alternatives with option value.Finally, we consider two frameworks that incorporate ambiguity aversion. We suggest that these results may be best used to guide future research aimed at improving the set of elicitation data.

In this article we ask, if quantities in an elicitation have been decomposed, is it better to combine experts before or after recomposing the quantities? We find that combining experts earlier, before recomposition of the quantities, leads to smaller errors with less variance. A simulation shows that these differences may be quite small on average; while an application to actual data shows that the differences can be significant in individual decision problems.

No abstract