Domain Adaptation Using Class Similarity for Robust Speech Recognition
This work addresses domain mismatch and data sparsity in speech recognition, offering an incremental improvement for robust acoustic modeling in specific adaptation tasks.
The paper tackles the challenge of domain adaptation for DNN acoustic models when target data is limited by proposing a method that transfers class similarity knowledge from source to target models, resulting in improved performance over fine-tuning, especially in highly mismatched domains like accent and noise adaptation.
When only limited target domain data is available, domain adaptation could be used to promote performance of deep neural network (DNN) acoustic model by leveraging well-trained source model and target domain data. However, suffering from domain mismatch and data sparsity, domain adaptation is very challenging. This paper proposes a novel adaptation method for DNN acoustic model using class similarity. Since the output distribution of DNN model contains the knowledge of similarity among classes, which is applicable to both source and target domain, it could be transferred from source to target model for the performance improvement. In our approach, we first compute the frame level posterior probabilities of source samples using source model. Then, for each class, probabilities of this class are used to compute a mean vector, which we refer to as mean soft labels. During adaptation, these mean soft labels are used in a regularization term to train the target model. Experiments showed that our approach outperforms fine-tuning using one-hot labels on both accent and noise adaptation task, especially when source and target domain are highly mismatched.