06-21 Uniform Multilingual Multi-Speaker Acoustic Model for Statistical Parametric Speech Synthesis of Low-Resourced Languages