This paper presents a solution to the problem of synthesising multilingual speech using waveform-concatenation speech synthesis. It presents methods for mapping the pronunciations of a target-language speaker onto the sounds available in the speech corpus of a native speaker so that the resulting synthesis produces speech which can accurately represent any foreign words encountered in a predominantly native-language text by use of multi-speaker synthesis. The methods differ depending on the language-pair and on the direction of the mapping, because in the case of one-to-many phonemic mappings, highlevel features can be used, but in the many-to-one case, a physical representation of the target speech signal is required. All mappings are automatic, and the use of rule-based procedures is minimised. In this way, the methods are extensible to any language combinations. Synthesised speech samples are included with the paper so that a subjective evaluation of the results can be made.
Cite as: Campbell, N. (2001) TALKING FOREIGN - concatenative speech synthesis and the language barrier. Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001), 337-340, doi: 10.21437/Eurospeech.2001-105
@inproceedings{campbell01_eurospeech,
author={Nick Campbell},
title={{TALKING FOREIGN - concatenative speech synthesis and the language barrier}},
year=2001,
booktitle={Proc. 7th European Conference on Speech Communication and Technology (Eurospeech 2001)},
pages={337--340},
doi={10.21437/Eurospeech.2001-105}
}