This paper presents a new baseline for Malay-English code switched speech corpus; which is constructed using a factored form of time delay neural networks (TDNN-F), which reflected a significant relative percentage reduction of 28.07% in the word-error rate (WER), as compared to the Gaussian Mixture Hidden Markov Models (GMM-HMMs). The presented results also confirm the effectiveness of time delay neural networks (TDNNs) for code-switched speech.
Cite as: Singh, A., Tan, T.-P. (2018) Evaluating Code-Switched Malay-English Speech Using Time Delay Neural Networks. Proc. 6th Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU 2018), 197-200, doi: 10.21437/SLTU.2018-41
@inproceedings{singh18_sltu, author={Anand Singh and Tien-Ping Tan}, title={{Evaluating Code-Switched Malay-English Speech Using Time Delay Neural Networks}}, year=2018, booktitle={Proc. 6th Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU 2018)}, pages={197--200}, doi={10.21437/SLTU.2018-41} }