Experiments on Speaker-Independent Phone Recognition Using BREF

A series of experiments for speaker-independent, continuous speech phone recognition have been carried out using the recently recorded BREF corpus. Our experiments are the rst to use this database, and are meant to provide a baseline performance evaluation for vocabulary independent phone recognition. The system was trained using hand-veriied data from 43 speakers. Using 35 context-independent phone models, a baseline phone accuracy of 60% (no phone grammar) has been obtained on an independent test set of 7635 phone segmentsfrom 19 speakers. Includingphone bigram probabilities as phonotactic constraints results in a performance of 63.5%. A phone accuracy of 68.6% (73.3 % correct) was obtained with 428 context dependent models.