Search

In this work, we studied the synthesis of Swiss German speech using different Text-to-Speech (TTS) models. We evaluated the TTS models on three corpora, and we found, that VITS models performed best, hence, using them for further testing. We also introduce a new method to evaluate TTS models by letting the discriminator of a trained vocoder GAN model predict whether a given waveform is human or synthesized. In summary, our best model delivers speech synthesis for different Swiss German dialects with previously unachieved quality.

We present the results and findings of the 2nd Swiss German speech to Standard German text shared task at SwissText 2022. Participants were asked to build a sentence-level Swiss German speech to Standard German text system specialized on the Grisons dialect. The objective was to maximize the BLEU score on a test set of Grisons speech. 3 teams participated, with the best-performing system achieving a BLEU score of 70.1., Comment: 3 pages, 0 figures, to appear in proceedings of SwissText 2022

We introduce Parallel Paraphrasing ($\text{Para}_\text{both}$), an augmentation method for translation metrics making use of automatic paraphrasing of both the reference and hypothesis. This method counteracts the typically misleading results of speech translation metrics such as WER, CER, and BLEU if only a single reference is available. We introduce two new datasets explicitly created to measure the quality of metrics intended to be applied to Swiss German speech-to-text systems. Based on these datasets, we show that we are able to significantly improve the correlation with human quality perception if our method is applied to commonly used metrics., Comment: Preprint SwissText 2023

In this paper, we present a learning rate method for gradient descent using only first order information. This method requires no manual tuning of the learning rate. We applied this method on a linear neural network built from scratch, along with the full-batch gradient descent, where we calculated the gradients for the whole dataset to perform one parameter update. We tested the method on a moderate sized dataset of housing information and compared the result with that of the Adam optimizer used with a sequential neural network model from Keras. The comparison shows that our method finds the minimum in a much fewer number of epochs than does Adam.

We present the Swiss Parliaments Corpus (SPC), an automatically aligned Swiss German speech to Standard German text corpus. This first version of the corpus is based on publicly available data of the Bernese cantonal parliament and consists of 293 hours of data. It was created using a novel forced sentence alignment procedure and an alignment quality estimator, which can be used to trade off corpus size and quality. We trained Automatic Speech Recognition (ASR) models as baselines on different subsets of the data and achieved a Word Error Rate (WER) of 0.278 and a BLEU score of 0.586 on the SPC test set. The corpus is freely available for download., Comment: 8 pages, 0 figures

The presented work focuses on the presentation of a discrete event simulator which can be used for automated sequencing and optimization of building processes. The sequencing is based on the commonly used component–activity–resource relations taking structural and process constraints into account. For the optimization a genetic algorithm approach was developed, implemented and successfully applied to several real life steel constructions. In this contribution we discuss the application of the discrete event simulator including its optimization capabilities on a 4D process model of a steel structure of an automobile recycling facility.

The invention relates to the use of L-xylulose as an inhibitor for saccharase and maltase enzyme activity.