End-to-End Code Switching Language Models for Automatic Speech Recognition
This addresses the challenge of processing code-switched speech for bilingual communities, but appears incremental as it builds on existing models like BERT without introducing a fundamentally new paradigm.
The paper tackles the problem of extracting monolingual text from code-switched speech in bilingual communities by proposing an approach using deep bidirectional language models like BERT and machine translation models, and explores methods for extracting code-switched text from ASR models, with results evaluated using metrics such as perplexity and WER compared to standard bilingual outputs.
In this paper, we particularly work on the code-switched text, one of the most common occurrences in the bilingual communities across the world. Due to the discrepancies in the extraction of code-switched text from an Automated Speech Recognition(ASR) module, and thereby extracting the monolingual text from the code-switched text, we propose an approach for extracting monolingual text using Deep Bi-directional Language Models(LM) such as BERT and other Machine Translation models, and also explore different ways of extracting code-switched text from the ASR model. We also explain the robustness of the model by comparing the results of Perplexity and other different metrics like WER, to the standard bi-lingual text output without any external information.