End-To-End Speech Synthesis Applied to Brazilian Portuguese

Voice synthesis systems have become increasingly popular in various applications, including personal assistants, GPS applications, screen readers and accessibility tools. Voice technology provides a natural way for human-computer interaction. However, not all languages have the same level of resources and systems available for voice synthesis. In this study, a team of researchers created publicly available resources for Brazilian Portuguese language in the form of a dataset and deep learning models for end-to-end voice synthesis. The dataset used in this study consists of 10.5 hours from a single speaker. The researchers investigated three different architectures to perform end-to-end speech synthesis: Tacotron 1, DCTTS and Mozilla TTS. They also analyzed the performance of models according to different vocoders (RTISI-LA, WaveRNN and Universal WaveRNN), phonetic transcriptions usage, transfer learning (from English) and denoising. In the proposed scenario, a model based on Mozilla TTS and RTISI-LA vocoder presented the best performance with a 4.03 MOS value. The researchers also found that transfer learning from English data, phonetic transcriptions usage and denoising are useful techniques to train models over the presented dataset. The obtained results are comparable to related works covering English even though they used a smaller dataset. This study's findings demonstrate that it is possible to create high-quality voice synthesis systems for Brazilian Portuguese using deep learning techniques with limited data availability. Overall, this research contributes significantly to improving voice technology's accessibility by providing publicly available resources for Brazilian Portuguese language that can be utilized in various applications such as personal assistants or accessibility tools.

Improving Voice Technology Accessibility with Deep Learning Techniques: A Study on Brazilian Portuguese

Voice technology has become increasingly popular in various applications, from personal assistants to GPS navigation and accessibility tools. It provides a natural way for human-computer interaction, but not all languages have the same level of resources and systems available for voice synthesis. To address this issue, a team of researchers created publicly available resources for Brazilian Portuguese language in the form of a dataset and deep learning models for end-to-end voice synthesis.

Dataset

The dataset used in this study consists of 10.5 hours from a single speaker. The data was collected using an open source speech collection toolkit called OpenSmile which records audio signals at 16kHz sampling rate and converts them into Mel spectrograms.

Models

The researchers investigated three different architectures to perform end-to-end speech synthesis: Tacotron 1, DCTTS and Mozilla TTS. They also analyzed the performance of models according to different vocoders (RTISI-LA, WaveRNN and Universal WaveRNN), phonetic transcriptions usage, transfer learning (from English) and denoising techniques. In the proposed scenario, they found that a model based on Mozilla TTS architecture combined with RTISI-LA vocoder presented the best performance with a 4.03 MOS value (Mean Opinion Score).

Findings

The researchers also found that transfer learning from English data, phonetic transcriptions usage and denoising are useful techniques to train models over the presented dataset. The obtained results were comparable to related works covering English even though they used a smaller dataset size than those studies did – demonstrating that it is possible to create high quality voice synthesis systems for Brazilian Portuguese using deep learning techniques with limited data availability.

Conclusion

This research contributes significantly to improving voice technology's accessibility by providing publicly available resources for Brazilian Portuguese language that can be utilized in various applications such as personal assistants or accessibility tools.