Attribute based shared hidden layers for cross-language knowledge transfer
Vipul Arora, Aditi Lahiri, Henning Reet · 2016
Deep neural network (DNN) acoustic models can be adapted to under-resourced languages by transferring the hidden layers. An analogous transfer problem is popular as few-shot learning to recognise scantily seen objects based on their meaningful attributes. In similar way, this paper proposes a principled way to represent the hidden layers of DNN in terms of attributes shared across languages. The diverse phoneme sets of different languages can be represented in terms of phonological features that are shared by them. The DNN layers estimating these features could then be transferred in a meaningful and reliable way. Here, we evaluate model transfer from English to German, by comparing the proposed method with other popular methods on the task of phoneme recognition. Experimental results support that apart from providing interpretability to the DNN acoustic models, the proposed framework provides efficient means for their speedy adaptation to different languages, even in the face of scanty adaptation data.