Corpus design for expressive speech: impact of the utterance length

“…To our knowledge, there is no publicly available French Tacotron2. Recent studies published on French synthesis focus on concatenation based TTS [17] or use Deep Convolutional TTS (DCTTS) [18]. DCTTS is a fully convolutional neural TTS, whose initial purpose was to alleviate the need for high computational power, while enabling quick training on smaller database.…”

Section: Related Workmentioning

confidence: 99%

Impact of Segmentation and Annotation in French end-to-end Synthesis

Lenglet¹,

Perrotin²,

Bailly³

2021

11th ISCA Speech Synthesis Workshop (SSW 11)

View full text Add to dashboard Cite

Audio books are commonly used to train text-to-speech models (TTS), as they offer large phonetic content with rather expressive pronunciation, but number and sizes of publicly available audio books corpora differ between languages. Moreover, the quality and accuracy of the available utterance segmentations are debatable. Yet, the impact of segmentation on the output synthesis is not well established. Additionally, utterances are generally used individually, without taking advantage of text level structuring information, even though they influence speaker reading. In this paper, we conduct a multidimensional evaluation of Tacotron2 trained on different segmentations and text level annotations of the same French corpus. We show that both spectrum accuracy and expressiveness depend on the segmentation used. In particular, a shorter segmentation, in addition with the annotation of paragraphs, benefits to spectrum reconstruction at the detriment of phrasing. Multidimensional analysis of mean opinion scores obtained with a MUSHRAexperiment revealed that phrasing was relatively more important than spectrum accuracy in perceptual judgement. This work serves as evidence that particular attention must be given to models evaluation, as well as how to use the training corpus to maximize synthesis characteristics of interest.

show abstract