Ninth International Conference on Spoken Language Processing

Pittsburgh, PA, USA
September 17-21, 2006

Evaluating Prosody of Mandarin Speech for Language Learning

Minghui Dong, Haizhou Li, Tin Lay Nwe

Institute for Infocomm Research, Singapore

This paper proposes an approach to automatically evaluate the prosody of Chinese Mandarin speech for language learning. In this approach, we grade the appropriateness of prosody of speech units according to a model speech corpus from a teacherís voice. To this end, we build two models, which are the prosody model and the scoring model. The prosody model that is built from the teacherís speech predicts the reference prosody for the learning text. The scoring model compares the studentís prosody with the reference prosody and gives a prosody rating score. Both the prosody model and the scoring model are built using regression tree. To make the two prosodies comparable, we transform the studentís prosody into the teacherís prosody space. To build the scoring model, we derive from the corpus a reference data set, in which prosody rating is associated with prosody parameters. During speech evaluation, the studentís prosody is first transformed into the teacherís prosody space and then evaluated by the scoring model. Experiments show that our model works well for speech of new speakers.

Full Paper

Bibliographic reference.  Dong, Minghui / Li, Haizhou / Nwe, Tin Lay (2006): "Evaluating prosody of Mandarin speech for language learning", In INTERSPEECH-2006, paper 1432-Wed3A3O.3.