Human language technologies for under-resourced African languages
: design, challenges, and prospects 1st ed.
Author
Additional Author(s)
Ekpenyong, Moses Effiong
Publisher
Cham, Switzerland : Springer International Publishing, 2018
Language
English
ISBN
9783319699608
Series
SpringerBriefs in speech technology, studies in speech signal processing, natural language understanding, and machine learning
Subject(s)
AUTOMATIC SPEECH RECOGNITION
NATURAL LANGUAGE PROCESSING (COMPUTER SCIENCE)
Notes
. .
Abstract
This book provides an overview of a recent and flexible approach to speech synthesis design to develop the first statistical parametric speech synthesizer for Ibibio, a West African tonal language. The design precludes the inflexibility encountered when modeling tonal features of the language and can be used for other tonal African languages. Mobile use and technological innovations in developing African nations have exploded. With mobile technology, many of the barriers caused by infrastructure issues have vanished. In order to address issues that are unique to African tonal languages, the book uses Ibibio as a model. The text reviews the language's speech characteristics, required for building the front end components of the design and propose a finite state transducer (FST), useful for modelling the language’s tonetactics. The statistical parametric approach discussed in the text, implements the Hidden Markov Model (HMM) technique, with the goal of creating a generic structure that learns the model from the text itself, and uses the data-driven approach to input specification. Provides a practical approach to tonal language synthesis that will broaden innovations in communications engineering Discusses a solution for tonal language system model design problems that are unique to growing markets in developing nations Addresses a growing technological need of developing countries language preservation projects.
