Enhancing and Adversarial: Improve ASR with Speaker Labels
This work addresses incremental improvements in ASR accuracy for speech processing applications by optimizing the use of speaker information in neural networks.
The paper tackles the problem of improving automatic speech recognition (ASR) by using speaker labels in multi-task learning, combining enhancing and adversarial training objectives. It achieves a 7% relative improvement on the Switchboard Hub5'00 set, matching the performance of i-vectors with adversarial training.
ASR can be improved by multi-task learning (MTL) with domain enhancing or domain adversarial training, which are two opposite objectives with the aim to increase/decrease domain variance towards domain-aware/agnostic ASR, respectively. In this work, we study how to best apply these two opposite objectives with speaker labels to improve conformer-based ASR. We also propose a novel adaptive gradient reversal layer for stable and effective adversarial training without tuning effort. Detailed analysis and experimental verification are conducted to show the optimal positions in the ASR neural network (NN) to apply speaker enhancing and adversarial training. We also explore their combination for further improvement, achieving the same performance as i-vectors plus adversarial training. Our best speaker-based MTL achieves 7\% relative improvement on the Switchboard Hub5'00 set. We also investigate the effect of such speaker-based MTL w.r.t. cleaner dataset and weaker ASR NN.