%0 Conference Paper %1 Liang2011 %A Liang, Hui %A Dines, John %B Proceedings of the 12th Annual Conference of the International Speech Communication Association (INTERSPEECH) %D 2011 %K HMM-based TTS, adaptation, cross-lingual error, generation knowledge minimum phonological speaker %P 1825-1828 %T Phonological Knowledge Guided HMM State Mapping for Cross-Lingual Speaker Adaptation %X Within the HMM state mapping-based cross-lingual speaker adaptation framework, the minimum Kullback-Leibler divergence criterion has been typically employed to measure the similarity of two average voice state distributions from two respective languages for state mapping construction. Considering that this simple criterion doesn't take any language-specific information into account, we propose a data-driven, phonological knowledge guided approach to strengthen the mapping construction -- state distributions from the two languages are clustered according to broad phonetic categories using decision trees and mapping rules are constructed only within each of the clusters. Objective evaluation of our proposed approach demonstrates reduction of mel-cepstral distortion and that mapping rules derived from a single training speaker generalize to other speakers, with subtle improvement being detected during subjective listening tests.

@inproceedings{Liang2011, abstract = {Within the HMM state mapping-based cross-lingual speaker adaptation framework, the minimum Kullback-Leibler divergence criterion has been typically employed to measure the similarity of two average voice state distributions from two respective languages for state mapping construction. Considering that this simple criterion doesn't take any language-specific information into account, we propose a data-driven, phonological knowledge guided approach to strengthen the mapping construction -- state distributions from the two languages are clustered according to broad phonetic categories using decision trees and mapping rules are constructed only within each of the clusters. Objective evaluation of our proposed approach demonstrates reduction of mel-cepstral distortion and that mapping rules derived from a single training speaker generalize to other speakers, with subtle improvement being detected during subjective listening tests.}, added-at = {2021-02-01T10:51:23.000+0100}, author = {Liang, Hui and Dines, John}, biburl = {https://www.bibsonomy.org/bibtex/28cf778f95549ec701d8dd71f7cdc0e36/m-toman}, booktitle = {Proceedings of the 12th Annual Conference of the International Speech Communication Association (INTERSPEECH)}, file = {:pdfs/liang_interspeech_2011.pdf:PDF}, interhash = {da70725ccad79e629e4f519784f13205}, intrahash = {8cf778f95549ec701d8dd71f7cdc0e36}, keywords = {HMM-based TTS, adaptation, cross-lingual error, generation knowledge minimum phonological speaker}, location = {Florence, Italy}, month = aug, owner = {mtoman}, pages = {1825-1828}, projects = {EMIME}, timestamp = {2021-02-01T10:51:23.000+0100}, title = {Phonological Knowledge Guided HMM State Mapping for Cross-Lingual Speaker Adaptation}, year = 2011 }