Multi-level adaptive networks in tandem and hybrid ASR systems

44 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Multi-Task Self-Supervised Learning for Robust Speech Recognition (2020, neutral): “Our approach is different from earlier DNN-based acoustic features extractors [17], as they exploited shared phonetic knowledge in a supervised manner, while PASE relies only on raw signals and self-supervised learning.”
Learning hidden unit contributions for unsupervised speaker adaptation of neural network acoustic models (2014, methodology): “In cases where speaker or channel adaptation can lead to significant improvements in accuracy, tandem systems often provide similar or reduced WERs compared with hybrid systems, owing to the applicability of GMM-based adaptation techniques [1, 9, 10].”
Transfer learning for speech and language processing (2015, methodology): “For example, in [73], the BN feature was first derived from a multilingual DNN, and then it was combined with the riginal feature to train a hybrid system.”
Regularization of context-dependent deep neural networks with context-independent multi-task training (2015, methodology): “As a second set of input features, we used tandem features obtained using the multi-level adaptive networks (MLAN) scheme [28].”