ISCA Archive Interspeech 2013
ISCA Archive Interspeech 2013

Annotation errors detection in TTS corpora

Jindřich Matoušek, Daniel Tihelka

We investigate the problem of automatic detection of annotation errors in single-speaker read-speech corpora used for text-to-speech (TTS) synthesis. Various word-level feature sets were used, and the performance of several detection methods based on support vector machines, extremely randomized trees, k-nearest neighbors, and the performance of novelty and outlier detection are evaluated. We show that both word- and utterance-level annotation error detections perform very well with both high precision and recall scores and with F1 measure being almost 90%, or 97%, respectively.


doi: 10.21437/Interspeech.2013-305

Cite as: Matoušek, J., Tihelka, D. (2013) Annotation errors detection in TTS corpora. Proc. Interspeech 2013, 1511-1515, doi: 10.21437/Interspeech.2013-305

@inproceedings{matousek13_interspeech,
  author={Jindřich Matoušek and Daniel Tihelka},
  title={{Annotation errors detection in TTS corpora}},
  year=2013,
  booktitle={Proc. Interspeech 2013},
  pages={1511--1515},
  doi={10.21437/Interspeech.2013-305}
}