Automatic measurement of vowel duration via structured prediction

TL;DR

This paper introduces a structured prediction machine learning model that automatically measures vowel duration from acoustic signals, outperforming traditional HMM-based forced aligners and reducing the need for manual annotation in phonetic studies.

Contribution

The paper presents a novel structured prediction approach for automatic vowel duration measurement that does not require phonetic transcription, improving accuracy over existing methods.

Findings

Model outperforms HMM-based forced aligners in accuracy

Requires no phonetic or orthographic transcription

Demonstrates scalability for phonetic research

Abstract

A key barrier to making phonetic studies scalable and replicable is the need to rely on subjective, manual annotation. To help meet this challenge, a machine learning algorithm was developed for automatic measurement of a widely used phonetic measure: vowel duration. Manually-annotated data were used to train a model that takes as input an arbitrary length segment of the acoustic signal containing a single vowel that is preceded and followed by consonants and outputs the duration of the vowel. The model is based on the structured prediction framework. The input signal and a hypothesized set of a vowel's onset and offset are mapped to an abstract vector space by a set of acoustic feature functions. The learning algorithm is trained in this space to minimize the difference in expectations between predicted and manually-measured vowel durations. The trained model can then automatically estimate vowel durations without phonetic or orthographic transcription. Results comparing the model to three sets of manually annotated data suggest it out-performed the current gold standard for duration measurement, an HMM-based forced aligner (which requires orthographic or phonetic transcription as an input).