EUROSPEECH 2001 Scandinavia
7th European Conference on Speech Communication and Technology
2nd INTERSPEECH Event

Aalborg, Denmark
September 3-7, 2001

                 

Is Non-Native Pronunciation Modelling Necessary ?

Silke Goronzy (1), Marina Sahakyan (1), Wolfgang Wokurek (2)

(1) Sony International (Europe), Germany
(2) IMS, University of Stuttgart, Germany

It is difficult to recognize non-native speech with speech recognition systems that are trained using native speech. While standard speaker adaptation techniques are often used in theses cases, they are not able to handle severe deviations from the expected pronunciation. Also, there has been a lot of interest in native pronunciation modelling recently. However, results often were not as good as expected. This paper investigates if a special treatment of non-native speakers is necessary. The effect of adding special pronunciation variants to the lexicon is examined. In contrast to native pronunciation modelling the results show that for the non-native case the enhanced dictionary is really necessary to obtain acceptable recognition rates. Recognition rates can be improved by up to 10% for German and even up to 28% for Italian learners of English. When combining this with MLLR adaptation, these results are further improved.

Full Paper

Bibliographic reference.  Goronzy, Silke / Sahakyan, Marina / Wokurek, Wolfgang (2001): "Is non-native pronunciation modelling necessary ?", In EUROSPEECH-2001, 309-312.