Neural Network Training Variations in Speech and Subsequent Performance Evaluation

Ewout van den Berg, Bhuvana Ramabhadran, Michael Picheny

Abstract:

In this work we study variance in the results of neural network training on a wide
variety of configurations in automatic speech recognition. Although this variance
itself is well known, this is, to the best of our knowledge, the first paper that
performs an extensive empirical study on its effects in speech recognition. We
view training as sampling from a distribution and show that these distributions
can have a substantial variance. These observations have important implications
on way results in the literature are reported and interpreted.

Conflicts: us.ibm.com