TY  - RPRT
TI  - Semi-supervised acoustic model training for five-lingual code-switched ASR
AU  - Astik Biswas
AU  - Emre Yılmaz
AU  - Febe de Wet
AU  - Ewald van der Westhuizen
AU  - Thomas Niesler
PY  - 2019
UR  - https://arxiv.org/abs/1906.08647
ID  - 1906.08647
ER  -