InterBiasing: Boost Unseen Word Recognition through Biasing Intermediate Predictions

TL;DR

This paper introduces InterBiasing, a parameter-free method using Self-conditioned CTC to enhance recognition of unseen words in speech recognition by biasing intermediate predictions with corrected labels, demonstrated on Japanese data.

Contribution

The paper proposes a novel, parameter-free approach that improves unseen word recognition by biasing intermediate predictions with corrected labels in a self-conditioned CTC framework.

Findings

Improved F1 score for unknown words in Japanese speech recognition

Effective biasing of intermediate predictions enhances recognition accuracy

Method does not require additional training parameters

Abstract

Despite recent advances in end-to-end speech recognition methods, their output is biased to the training data's vocabulary, resulting in inaccurate recognition of unknown terms or proper nouns. To improve the recognition accuracy for a given set of such terms, we propose an adaptation parameter-free approach based on Self-conditioned CTC. Our method improves the recognition accuracy of misrecognized target keywords by substituting their intermediate CTC predictions with corrected labels, which are then passed on to the subsequent layers. First, we create pairs of correct labels and recognition error instances for a keyword list using Text-to-Speech and a recognition model. We use these pairs to replace intermediate prediction errors by the labels. Conditioning the subsequent layers of the encoder on the labels, it is possible to acoustically evaluate the target keywords. Experiments conducted in Japanese demonstrated that our method successfully improved the F1 score for unknown words.