5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Use of Recurrent Network for Unknown Language Rejection in Language Identification System

HingKeung Kwan, Keikichi Hirose

Dept. of Information and Communication Engineering, University of Tokyo, Japan

In the past, we attempted to use a multilayer perceptron neural network as a means to prevent those unknown language inputs from being misidentified as one of the target languages in language identification system. However, the use of multilayer perceptron neural network could not utilize the temporal information from the utterances. Results show that with the use of phonemic unigram as input features to a recurrent neural network of Jordan architecture, a 3 target language identification rate of 98.1% can be achieved. By setting the output thresholds to 0.6 to reject 2 more unknown languages, a lower overall rate of 85.9% is obtained.

Full Paper

Bibliographic reference.  Kwan, HingKeung / Hirose, Keikichi (1997): "Use of recurrent network for unknown language rejection in language identification system", In EUROSPEECH-1997, 63-66.