Noise Robust Speech Recognition Based on Noise-Adapted HMMs Using Speech Feature Compensation

In conventional VTS-based noisy speech recognition methods, the parameters of the clean HMM are adapted to test noisy speech, or the original clean speech is estimated from the test noisy speech. However, in noisy speech recognition, improved performance is generally expected by employing noisy acoustic models produced by methods such as MTR and MMSR compared with using clean HMMs. In this research, a method was devised that can make use of the noisy acoustic models in the conventional VTS algorithm. A novel mathematical relation was derived between the test and training noisy speech and MMSE of the training noisy speech is obtained from the test noisy speech based on the relation. The proposed method was applied to noise-adapted HMMs trained by the MTR and MMSR and could reduce the relative word error rate by 6.5% and 7.2%, respectively, in the noisy speech recognition experiments on the Aurora 2 database.