When CTC Training Meets Acoustic Landmarks

TL;DR

This paper enhances CTC training for acoustic models by integrating acoustic landmarks, leading to faster convergence and lower error rates across multiple speech recognition datasets.

Contribution

It introduces a novel set of acoustic landmarks and mixed label sequences to improve CTC convergence and accuracy, demonstrating effectiveness on TIMIT and WSJ datasets.

Findings

Faster and smoother CTC convergence with landmarks

Reduced recognition error rates by 8.72% on TIMIT

First verification of acoustic landmark theory on a mid-sized ASR task

Abstract

Connectionist temporal classification (CTC) provides an end-to-end acoustic model (AM) training strategy. CTC learns accurate AMs without time-aligned phonetic transcription, but sometimes fails to converge, especially in resource-constrained scenarios. In this paper, the convergence properties of CTC are improved by incorporating acoustic landmarks. We tailored a new set of acoustic landmarks to help CTC training converge more rapidly and smoothly while also reducing recognition error rates. We leveraged new target label sequences mixed with both phone and manner changes to guide CTC training. Experiments on TIMIT demonstrated that CTC based acoustic models converge significantly faster and smoother when they are augmented by acoustic landmarks. The models pretrained with mixed target labels can be further finetuned, resulting in phone error rates 8.72% below baseline on TIMIT. Consistent performance gain is also observed on WSJ (a larger corpus) and reduced TIMIT (smaller). With WSJ, we are the first to succeed in verifying the effectiveness of acoustic landmark theory on a mid-sized ASR task.