Attentive Temporal Pooling for Conformer-based Streaming Language Identification in Long-form Speech

Abstract

In this paper, we introduce a novel language identification system based on conformer layers. We propose an attentive temporal pooling mechanism to allow the model to carry information in long-form audio via a recurrent form, such that the inference can be performed in a streaming fashion. Additionally, we investigate two domain adaptation approaches to allow adapting an existing language identification model without retraining the model parameters for a new domain. We perform a comparative study of different model topologies under different constraints of model size, and find that conformer-based models significantly outperform LSTM and transformer based models. Our experiments also show that attentive temporal pooling and domain adaptation improve model accuracy.

Full paper on arXiv

Details

arXiv identifier

2202.12163

Published

2022-02-24

Authors

Quan Wang, Yang Yu, Jason Pelecanos, Yiling Huang, Ignacio Lopez Moreno

Open Models Built on This Paper

Every model in the SAVRN Model Hub whose card cites this paper, most downloaded first, with what it takes to run each one.