We describe a novel method for tuning the decoding parameters of a speech-to-text system so as to minimize word error rate (WER) subject to an over-all time constraint. When applied to three sub-realtime systems for recognizing English conversational telephone speech, the method gave speed improvements of up to 21.1% while at the same time reducing WER by up to 6.7%.
Bibliographic reference. Colthurst, Thomas / Arvizo, Tresi / Kao, Chia-Lin / Kimball, Owen / Lowe, Stephen A. / Miller, David R. H. / Sciver, Jim Van (2007): "Parameter tuning for fast speech recognition", In INTERSPEECH-2007, 1477-1480.