Text Preprocessing for Speech Synthesis

Uwe D. Reichel, Hartmut R. Pfitzinger

Open access LMU (Ludwid Maxmilian's Universitat Munchen) · 2006 · 18 citations · 12 references

DOIFull text

Open access

Concepts

Abstract

In this paper we describe our text preprocessing modules for English text-to-speech synthesis.These modules comprise rule-based text normalization subsuming sentence segmentation and normalization of non-standard words, statistical part-of-speech tagging, and statistical syllabification, grapheme-to-phoneme conversion, and word stress assignment relying in parts on rule-based morphological analysis.

References

12