Leveraging Phone Mask Training for Phonetic-Reduction-Robust E2E Uyghur Speech Recognition

TL;DR

This paper introduces a phone mask training technique for Uyghur speech recognition that improves accuracy by simulating phonetic reduction phenomena, especially in spontaneous speech, through strategic masking during model training.

Contribution

The study proposes a novel phone mask training method for Conformer-based Uyghur E2E speech recognition, enhancing robustness against phonetic reduction effects in spontaneous speech.

Findings

Achieved 5.51% relative WER reduction on reading speech.

Achieved 12.92% relative WER reduction on spontaneous speech.

Demonstrated 20% relative improvement on open-source dataset THUYG-20.

Abstract

In Uyghur speech, consonant and vowel reduction are often encountered, especially in spontaneous speech with high speech rate, which will cause a degradation of speech recognition performance. To solve this problem, we propose an effective phone mask training method for Conformer-based Uyghur end-to-end (E2E) speech recognition. The idea is to randomly mask off a certain percentage features of phones during model training, which simulates the above verbal phenomena and facilitates E2E model to learn more contextual information. According to experiments, the above issues can be greatly alleviated. In addition, deep investigations are carried out into different units in masking, which shows the effectiveness of our proposed masking unit. We also further study the masking method and optimize filling strategy of phone mask. Finally, compared with Conformer-based E2E baseline without mask training, our model demonstrates about 5.51% relative Word Error Rate (WER) reduction on reading speech and 12.92% on spontaneous speech, respectively. The above approach has also been verified on test-set of open-source data THUYG-20, which shows 20% relative improvements.