Development of dialect-specific speech recognizers using adaptation methods

Several adaptation approaches have been proposed in an e ort to improve the speech recognition performance in mismatched conditions. However, the application of these approaches had been mostly constrained to the speaker or channel adaptation tasks. In this paper, we rst investigate the e ect of mismatched dialects between training and testing speakers in an Automatic Speech Recognition (ASR) system. We nd that a mismatch in dialects significantly in uences the recognition accuracy. Consequently, we apply several adaptation approaches to develop a dialect-speci c recognition system using a dialect-dependent system trained on a di erent dialect and a small number of training sentences from the target dialect. We show that adaptation improves recognition performance dramatically with small amounts of training sentences. We further show that, although the recognition performance of traditionally trained systems highly degrades as we decrease the number of training speakers, the performance of adapted systems is not in uenced so much.