8th International Conference on Spoken Language Processing

Jeju Island, Korea
October 4-8, 2004

Cross-Lingual Phoneme Mapping for Multilingual Synthesis Systems

Marko Moberg, Kimmo Parssinen, Juha Iso-Sipila

Nokia, Finland

Development of a multilingual text-to-speech (TTS) system requires usually a lot of time, effort and language resources. The implementation tends to consume large amounts of memory as the number of supported languages increases. This paper proposes a simple method for quickly increasing the language portfolio of an existing TTS system with the minimal effort and memory consumption. The cross-lingual phoneme mapping modifies the phonetic transcription of a new language by presenting it with the phoneme set supported by an existing TTS system. The synthesis output of the mapped language is quite accurate phonetically but the intonation is based on the existing target language. The evaluations showed that the cross-lingual phoneme mapping could provide an adequate quality synthesis when the language portfolio must be rapidly expanded and the memory and the language resources are scarce. The method is best applied for the synthesis of short utterances including names and isolated words.

Full Paper

Bibliographic reference.  Moberg, Marko / Parssinen, Kimmo / Iso-Sipila, Juha (2004): "Cross-lingual phoneme mapping for multilingual synthesis systems", In INTERSPEECH-2004, 1029-1032.