Sixth European Conference on Speech Communication and Technology

Budapest, Hungary
September 5-9, 1999

Selection of Waveform Units for Corpus-Based Mandarin Speech Synthesis Based on Decision Trees and Prosodic Modification Costs

Fu-chiang Chou (1), Chiu-yu Tseng (1), Lin-shan Lee (2)

(1) Institute of Linguistics, Academia Sinica; (2) Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan

A lazy decision tree approach is described in this paper for the selection of concatenative units for Mandarin speech synthesis. The concept is not to induce a concise hypothesis from a given training data; the selection is delayed until a test instance is given. Thus we can construct the "best" decision tree for each selection. The selection of waveform units is guided with other simultaneously selected prosodic parameters. The selected waveform units can be directly concatenated into output speech or modified with the selected prosodic parameters. This method can produce synthetic speech that sounds very natural and resembles the acoustic and prosodic characteristics of the original speaker. A Mandarin speech synthesizer is described in this paper. However, most of the technologies are language independent and can be extended to a multi-lingual system.

Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Chou, Fu-chiang / Tseng, Chiu-yu / Lee, Lin-shan (1999): "Selection of waveform units for corpus-based Mandarin speech synthesis based on decision trees and prosodic modification costs", In EUROSPEECH'99, 2295-2298.