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.

Respiratory insufficiency is a symptom that requires hospitalization. This work investigates whether it is possible to detect this condition by analyzing patient's speech samples; the analysis was performed on data collected during the first wave of the COVID-19 pandemic in 2020, and thus limited to respiratory insufficiency in COVID-19 patients. For that, a dataset was created consisting of speech emissions of both COVID-19 patients affected by respiratory insufficiency and a control group. This dataset was used to build a Convolution Neural Network to detect respiratory insufficiency using speech emission MFCC representations. Methodologically, dealing with background noise was a challenge, so we also collected background noise from COVID-19 wards where patients were located. Due to the difficulty in filtering noise without eliminating crucial information, noise samples were injected in the control group data to prevent bias. Moreover, we investigated (i) two approaches to address the duration variance of audios, and (ii) the ideal number of noise samples to inject in both patients and the control group to prevent bias and overfitting. The techniques developed reached 91.66% accuracy. Thus we validated the project's Leading Hypothesis, namely that it is possible to detect respiratory insufficiency in speech utterances, under real-life environmental conditions; we believe our results justify further enquiries into the use of automated speech analysis to support health professionals in triage procedures.

In this work, we propose several techniques to address data scarceness in ComParE 2021 COVID-19 identification tasks for the application of deep models such as Convolutional Neural Networks. Data is initially preprocessed into spectrogram or MFCC-gram formats. After preprocessing, we combine three different data augmentation techniques to be applied in model training. Then we employ transfer learning techniques from pretrained audio neural networks. Those techniques are applied to several distinct neural architectures. For COVID-19 identification in speech segments, we obtained competitive results. On the other hand, in the identification task based on cough data, we succeeded in producing a noticeable improvement on existing baselines, reaching 75.9% unweighted average recall (UAR).

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 zeroshot multi-speaker and multilingual training. We achieved stateof-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 zeroshot 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.

In this paper, we propose SC-GlowTTS: an efficient zero-shot multi-speaker text-to-speech model that improves similarity for speakers unseen in training. We propose a speaker-conditional architecture that explores a flow-based decoder that works in a zero-shot scenario. As text encoders, we explore a dilated residual convolutional-based encoder, gated convolutional-based encoder, and transformer-based encoder. Additionally, we have shown that adjusting a GAN-based vocoder for the spectrograms predicted by the TTS model on the training dataset can significantly improve the similarity and speech quality for new speakers. Our model is able to converge in training, using only 11 speakers, reaching state-of-the-art results for similarity with new speakers, as well as high speech quality.