Computational Pronunciation Analysis in Sung Utterances
This work addresses pronunciation modeling for automatic lyrics transcription, an incremental improvement in a domain-specific task.
The paper tackled the problem of pronunciation variances in sung utterances for automatic lyrics transcription by proposing a singing-adapted pronunciation model, which outperformed standard speech dictionaries in word recognition experiments on multiple datasets, achieving the best results on a capella recordings with n-gram language models.
Recent automatic lyrics transcription (ALT) approaches focus on building stronger acoustic models or in-domain language models, while the pronunciation aspect is seldom touched upon. This paper applies a novel computational analysis on the pronunciation variances in sung utterances and further proposes a new pronunciation model adapted for singing. The singing-adapted model is tested on multiple public datasets via word recognition experiments. It performs better than the standard speech dictionary in all settings reporting the best results on ALT in a capella recordings using n-gram language models. For reproducibility, we share the sentence-level annotations used in testing, providing a new benchmark evaluation set for ALT.