Location, Location: Enhancing the Evaluation of Text-to-Speech Synthesis Using the Rapid Prosody Transcription Paradigm

TL;DR

This paper introduces a novel evaluation method for text-to-speech systems using the Rapid Prosody Transcription paradigm, enabling detailed error localization and analysis of prosodic features beyond traditional MOS scores.

Contribution

It proposes a new real-time error marking approach for TTS evaluation that provides detailed insights into prosodic errors and their relation to system performance.

Findings

Error marks cluster around prosodic boundaries in audiobook samples.

The method correlates with MOS-based rankings but offers more detailed error localization.

Differences in prosody generation are observed across TTS systems in question-answer stimuli.

Abstract

Text-to-Speech synthesis systems are generally evaluated using Mean Opinion Score (MOS) tests, where listeners score samples of synthetic speech on a Likert scale. A major drawback of MOS tests is that they only offer a general measure of overall quality-i.e., the naturalness of an utterance-and so cannot tell us where exactly synthesis errors occur. This can make evaluation of the appropriateness of prosodic variation within utterances inconclusive. To address this, we propose a novel evaluation method based on the Rapid Prosody Transcription paradigm. This allows listeners to mark the locations of errors in an utterance in real-time, providing a probabilistic representation of the perceptual errors that occur in the synthetic signal. We conduct experiments that confirm that the fine-grained evaluation can be mapped to system rankings of standard MOS tests, but the error marking gives a much more comprehensive assessment of synthesized prosody. In particular, for standard audiobook test set samples, we see that error marks consistently cluster around words at major prosodic boundaries indicated by punctuation. However, for question-answer based stimuli, where we control information structure, we see differences emerge in the ability of neural TTS systems to generate context-appropriate prosodic prominence.