Enforcing Encoder-Decoder Modularity in Sequence-to-Sequence Models

11/09/2019

∙

by   Siddharth Dalmia, et al.

∙

0

∙

share

Inspired by modular software design principles of independence, interchangeability, and clarity of interface, we introduce a method for enforcing encoder-decoder modularity in seq2seq models without sacrificing the overall model quality or its full differentiability. We discretize the encoder output units into a predefined interpretable vocabulary space using the Connectionist Temporal Classification (CTC) loss. Our modular systems achieve near SOTA performance on the 300h Switchboard benchmark, with WER of 8.3 17.6 modules which are independent and interchangeable.