Mixture Pruning and Roughening for Scalable Acoustic ModelsComputer Science Department
Date of Original Version1-1-2008
Abstract or DescriptionIn an automatic speech recognition system using a tied-mixture acoustic model, the main cost in CPU time and memory lies not in the evaluation and storage of Gaussians themselves but rather in evaluating the mixture likelihoods for each state output distribution. Using a simple entropy-based technique for pruning the mixture weight distributions, we can achieve a signiﬁcant speedup in recognition for a 5000-word vocabulary with a negligible increase in word error rate. This allows us to achieve real-time connected-word dictation on an ARM-based mobile device.
Citation InformationAlexander I Rudnicky and David Huggins-Daines. "Mixture Pruning and Roughening for Scalable Acoustic Models" (2008)
Available at: http://works.bepress.com/alexander_rudnicky/69/