Reproduced some classic and state-of-the-art KWS model in Pytorch or Pytorch-Kaldi, including CNN, LSTM, TDNN, DSCNN and ResNet.
Reproduced Google's generalized end-to-end loss in speaker verification.
[1] Sainath, T. N., & Parada, C. (2015). Convolutional neural networks for small-footprint keyword spotting. In Sixteenth Annual Conference of the International Speech Communication Association.
[2] Tang, R., & Lin, J. (2018, April). Deep residual learning for small-footprint keyword spotting. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 5484-5488). IEEE.
[3] Myer, S., & Tomar, V. S. (2018). Efficient keyword spotting using time delay neural networks. arXiv preprint arXiv:1807.04353.
[4] Chen, G., Parada, C., & Sainath, T. N. (2015, April). Query-by-example keyword spotting using long short-term memory networks. In 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 5236-5240). IEEE.
[5] Zhang, Y., Suda, N., Lai, L., & Chandra, V. (2017). Hello edge: Keyword spotting on microcontrollers. arXiv preprint arXiv:1711.07128.
[6] Ma, H., Bai, Y., Yi, J., & Tao, J. (2019, November). Hypersphere Embedding and Additive Margin for Query-by-example Keyword Spotting. In 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) (pp. 868-872). IEEE.
[7] Audhkhasi, K., Rosenberg, A., Sethy, A., Ramabhadran, B., & Kingsbury, B. (2017). End-to-end ASR-free keyword search from speech. IEEE Journal of Selected Topics in Signal Processing, 11(8), 1351-1359.
[8] Wan, L., Wang, Q., Papir, A., & Moreno, I. L. (2018, April). Generalized end-to-end loss for speaker verification. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 4879-4883). IEEE.