Code-Switching Detection with Data-Augmented Acoustic and Language Models

Emre Yilmaz, Henk Van Den Heuvel


In this paper, we investigate the code-switching detection performance of a code-switching (CS) automatic speech recognition (ASR) system with data-augmented acoustic and language models. We focus on the recognition of Frisian-Dutch radio broadcasts where one of the mixed languages, namely Frisian, is under-resourced. Recently, we have explored how the acoustic modeling AM) can benefit from monolingual speech data belonging to the high-resourced mixed language. For this purpose, we have trained state-of-the-art AMs on a significantly increased amount of CS speech by applying automatic transcription and monolingual Dutch speech. Moreover, we have improved the language model (LM) by creating CS text in various ways including text enervation using recurrent LMs trained on existing CS text. Motivated by the significantly improved CS ASR performance, we delve into the CS detection performance of the same ASR system in this work by reporting CS detection accuracies together with a detailed detection error analysis.


 DOI: 10.21437/SLTU.2018-27

Cite as: Yilmaz, E., Van Den Heuvel, H. (2018) Code-Switching Detection with Data-Augmented Acoustic and Language Models. Proc. The 6th Intl. Workshop on Spoken Language Technologies for Under-Resourced Languages, 127-131, DOI: 10.21437/SLTU.2018-27.


@inproceedings{Yilmaz2018,
  author={Emre Yilmaz and  Henk {Van Den Heuvel} and David {Van Leeuwen}},
  title={{Code-Switching Detection with Data-Augmented Acoustic and Language Models}},
  year=2018,
  booktitle={Proc. The 6th Intl. Workshop on Spoken Language Technologies for Under-Resourced Languages},
  pages={127--131},
  doi={10.21437/SLTU.2018-27},
  url={http://dx.doi.org/10.21437/SLTU.2018-27}
}