arXiv subjects

Julian Weber

Publications and source records attributed to Julian Weber.

5 recordsLinked to original sources

arXiv2024

XTTS: a Massively Multilingual Zero-Shot Text-to-Speech Model

Most Zero-shot Multi-speaker TTS (ZS-TTS) systems support only a single language. Although models like YourTTS, VALL-E X, Mega-TTS 2, and Voicebox explored Multilingual ZS-TTS they are limited to just a few high/medium resource languages, limiting the applications of these models in most of the low/medium resource languages. In this paper, we aim to alleviate this issue by proposing and making publicly available the XTTS system. Our method builds upon the Tortoise model and adds several novel modifications to enable multilingual training, improve voice cloning, and enable faster training and inference. XTTS was trained in 16 languages and achieved state-of-the-art (SOTA) results in most of them.

eess.AS↗

arXiv2022

Interactive Process Identification and Selection from SAP ERP

SAP ERP is one of the most popular information systems supporting various organizational processes, e.g., O2C and P2P. However, the amount of processes and data contained in SAP ERP is enormous. Thus, the identification of the processes that are contained in a specific SAP instance, and the creation of a list of related tables is a significant challenge. Eventually, one needs to extract an event log for process mining purposes from SAP ERP. This demo paper shows the tool Interactive SAP Explorer that tackles the process identification and selection problem by encoding the relational structure of SAP ERP in a labeled property graph. Our approach allows asking complex process-related queries along with advanced representations of the relational structure.

cs.DB↗

arXiv2021

YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone

YourTTS brings the power of a multilingual approach to the task of zero-shot multi-speaker TTS. Our method builds upon the VITS model and adds several novel modifications for zero-shot multi-speaker and multilingual training. We achieved state-of-the-art (SOTA) results in zero-shot multi-speaker TTS and results comparable to SOTA in zero-shot voice conversion on the VCTK dataset. Additionally, our approach achieves promising results in a target language with a single-speaker dataset, opening possibilities for zero-shot multi-speaker TTS and zero-shot voice conversion systems in low-resource languages. Finally, it is possible to fine-tune the YourTTS model with less than 1 minute of speech and achieve state-of-the-art results in voice similarity and with reasonable quality. This is important to allow synthesis for speakers with a very different voice or recording characteristics from those seen during training.

cs.SD↗

arXiv2017

Large scale Micro-Photometry for high resolution pH-characterization during electro-osmotic pumping and modular micro-swimming

Micro-fluidic pumps as well as artificial micro-swimmers are conveniently realized exploiting phoretic solvent flows based on local gradients of temperature, electrolyte concentration or pH. We here present a facile micro-photometric method for monitoring pH gradients and demonstrate its performance and scope on different experimental situations including an electro-osmotic pump and modular micro-swimmers assembled from ion exchange resin beads and polystyrene colloids. In combination with the present microscope and DSLR camera our method offers a 2 μm spatial resolution at video frame rate over a field of view of 3920x2602 μm^2. Under optimal conditions we achieve a pH-resolution of 0.05 with about equal contributions from statistical and systematical uncertainties. Our quantitative micro-photometric characterization of pH gradients which develop in time and reach out several mm is anticipated to provide valuable input for reliable modeling and simulations of a large variety of complex flow situations involving pH-gradients including artificial micro-swimmers, microfluidic pumping or even electro-convection.

cond-mat.soft↗

arXiv2017

Assembly and speed control in ion exchange based modular phoretic micro-swimmers

We report an experimental study on ion-exchange based modular micro-swimmers in low-salt water. Cationic ion-exchange particles and passive cargo particles assemble into self-propelling complexes, showing self-propulsion at speeds of several microns per second over extended distances and times. We quantify the assembly and speed of the complexes for different combinations of ion exchange particles and cargo particles, substrate types, salt types and concentrations, and cell geometries. Irrespective of experimental boundary conditions, we observe a regular development of the assembly shape with increasing number of cargo. Moreover, the swimming speed increases stepwise upon increasing the number of cargo and then saturates at a maximum speed, indicating an active role of cargo in modular swimming. We propose a geometric model of self-assembly to describe the experimental observations in a qualitative way. Our study also provides some constraints for future theoretical modelling and simulation.

cond-mat.soft↗