Enhancing the recognition of children's speech on acoustically mismatched ASR system
Syed Shahnawazuddin, Hemant Kumar Kathania, Rohit Sinha · 2015
The work presented in this paper explores the issues of recognizing children's speech using acoustic models trained on adults' speech data. In such conditions, on account of large acoustic mismatch between training and test data, a high degradation in the recognition performance is noted. In our earlier work, a binary weighting of cepstral features as well as of acoustic model parameters was explored to address the same. In this paper, a soft-weighting is proposed to overcome the information loss with simple binary weighting scheme. This is achieved through a low-rank projection learned using adults' training data. The so derived transform happens to emphasize the principal dimensions of acoustic variations in adults' speech. During testing, the transform maps children's test data to the space of the training data and thus suppresses the mismatched dimensions. The proposed scheme is also verified experimentally using a recognition system trained on adults' data only as well as another system trained using adults' and children's data pooled together. The effectiveness of acoustic model adaptation is also explored to further enhance the system performance. Combining SW with cluster model interpolation leads to a relative improvement of 14% over the baseline.