
Ultrasound-Based Silent Speech Interface Using Convolutional and Recurrent Neural Networks
Silent Speech Interface (SSI) is a technology with the goal of synthesizing speech from articulatory motion. A Deep Neural Network based SSI using ultrasound images of the tongue as input signals and spectral coefficients of a vocoder as target parameters are proposed. Several deep
learning models, such as a baseline Feed-forward, and a combination of Convolutional and Recurrent Neural Networks are presented and discussed. A pre-processing step using a Deep Convolutional AutoEncoder was also studied. According to the experimental results, an architecture based on a CNN
and bidirectional LSTM layers has shown the best objective and subjective results.
7 References.
No Supplementary Data.
No Article Media
No Metrics
Document Type: Research Article
Publication date: July 1, 2019
This article was made available online on July 4, 2019 as a Fast Track article with title: "Ultrasound-based silent speech interface using convolutional and recurrent neural networks".
- Acta Acustica united with Acustica, published together with the European Acoustics Association (EAA), is an international, peer-reviewed journal on acoustics. It publishes original articles on all subjects in the field of acoustics, such as general linear acoustics, nonlinear acoustics, macrosonics, flow acoustics, atmospheric sound, underwater sound, ultrasonics, physical acoustics, structural acoustics, noise control, active control, environmental noise, building acoustics, room acoustics, acoustic materials, acoustic signal processing, computational and numerical acoustics, hearing, audiology and psychoacoustics, speech, musical acoustics, electroacoustics, auditory quality of systems. It reports on original scientific research in acoustics and on engineering applications. The journal considers scientific papers, technical and applied papers, book reviews, short communications, doctoral thesis abstracts, etc. In irregular intervals also special issues and review articles are published.
- Editorial Board
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Information for Advertisers
- Online User License
- Ingenta Connect is not responsible for the content or availability of external websites