CSDL Bài trích Báo - Tạp chí
chủ đề: Engineering
1 ADAPT-TTS: high-quality zero-shot multi-speaker text-to-speech adaptive-based for Vietnamese / Phuong Pham Ngoc, Chung Tran Quang, Mai Luong Chi // .- 2023 .- Tập 39 - Số 2 .- P. 159-173 .- 621
In this paper introduce the Adapt-TTS model that allows high-quality audio synthesis from a small adaptive sample without training to solve these problems. Key recommendations: 1. The Extracting Mel-vector (EMV) architecture allows for a better representation of speaker characteristics and speech style; 2. An improved zero-shot model with a denoising diffusion model (Mel-spectrogram denoiser) component allows for new voice synthesis without training with better quality (less noise).
2 Artificial neural network for regression analysis developed in excel VBA: a case study in pile bearing capacity prediction / Hoang Nhat Đuc // Khoa học & Công nghệ Đại học Duy Tân .- 2022 .- No.6(55) .- P. 16-22 .- 624
This paper aims at developing an artificial neural network (ANN) for regression analysis. The ANN model is developed in Excel Visual Basic for Applications (VBA) to facilitate its practical implementations. The capability of the developed ANN program has been tested with the task of pile bearing capacity prediction.
3 Image processing-based measurement of pavement macrotexture depth with sand patch test / Hoang Nhat Đuc, Tran Van Đuc // Khoa học & Công nghệ Đại học Duy Tân .- 2022 .- No. 6(55) .- P. 54-58 .- 624
This research work utilizes image processing techniques and sand patch tests to measure pavement macrotexture Image processing techniques including median filter, Sobel edge detector, and circle fitting are used. The newly developed method is able to compute the pavement macrotexture depth index based on image of sand patch test. This image processing-based method is developed in Visual C#.NET framework 4.7.2. The newly developed program has been tested with real-world image samples.