Learning discriminative features in sequence training without requiring framewise labelled data
This addresses the challenge of acoustic variability in ASR for industrial applications without requiring presegmented data, offering a practical improvement over existing methods.
The paper tackled the problem of learning discriminative features for automatic speech recognition without needing framewise labeled data, and the result was a novel method that consistently outperformed state-of-the-art models, reducing Word Error Rate by up to 12.94% under unseen noise conditions.
In this work, we try to answer two questions: Can deeply learned features with discriminative power benefit an ASR system's robustness to acoustic variability? And how to learn them without requiring framewise labelled sequence training data? As existing methods usually require knowing where the labels occur in the input sequence, they have so far been limited to many real-world sequence learning tasks. We propose a novel method which simultaneously models both the sequence discriminative training and the feature discriminative learning within a single network architecture, so that it can learn discriminative deep features in sequence training that obviates the need for presegmented training data. Our experiment in a realistic industrial ASR task shows that, without requiring any specific fine-tuning or additional complexity, our proposed models have consistently outperformed state-of-the-art models and significantly reduced Word Error Rate (WER) under all test conditions, and especially with highest improvements under unseen noise conditions, by relative 12.94%, 8.66% and 5.80%, showing our proposed models can generalize better to acoustic variability.