IVN-Based Joint Training Of GMM And HMMs Using An Improved VTS-Based Feature Compensation For Noisy Speech Recognition.

Jun Du, Qiang Huo

Browse the full Interspeech paper archive.