Yue Phonetic Database
Covers over 300,000 Cantonese entries
Includes standard Jyutping romanization
All major romanization systems can be provided
Overview
CJKI’s Yue Phonetic Database (YPD) provides Cantonese readings for 300,000 compound words and approximately 80,000 readings and romanized variants for about 13,000 single Traditional Chinese characters.
YPD features phonemic transcriptions given in the standard Jyutping romanization, also available in up to ten Cantonese romanization systems (IPA accurate transcriptions also possible). The readings are ordered by frequency and/or importance, while flags distinguish common readings from rare ones.
Practical Applications
YPD is ideal for applications such as:
Natural language processing applications
Such as speech recognition and speech synthesis
Machine translation
For use in input method editors, TTS for car navigation systems and speech-to-speech systems
Reference Documents
Related Resources

Chinese Phonetic Database
Phonemic transcriptions showing differences between PRC and Taiwan

Japanese Phonetic Database
IPA phonetic and phonemic transcriptions for core Japanese vocabulary

Chinese Hanyu Pinyin Database
Accurate hanyu pinyin data including technical terms and proper nouns