CNN-based phone segmentation experiments in a less-represented language - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

CNN-based phone segmentation experiments in a less-represented language

Résumé

These last years, there has been a regain of interest in unsupervised sub-lexical and lexical unit discovery. Speech segmentation into phone-like units may be a first interesting step for such a task. In this article, we report speech segmentation experiments in Xitsonga, a less-represented language spoken in South Africa. We chose to use convolutional neural networks (CNN) with FBANK static coefficients as input. The models take binary decisions whether a boundary is present or not at each signal sliding frame. We compare the use of a model trained exclusively on Xitsonga data to the use of a bootstrap model trained on a larger corpus of another language, the BUCKEYE U.S. English corpus. Using a two-convolution-layer model, a 79% F-measure was obtained on BUCKEYE, with a 20 ms error tolerance. This performance is equal to the human inter-annotator agreement rate. We then used this bootstrap model to segment Xitsonga data and compared the results when adapting it with 1 to 20 minutes of Xitsonga data.
Fichier principal
Vignette du fichier
manenti_17052.pdf (556.23 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01500519 , version 1 (03-04-2017)

Identifiants

  • HAL Id : hal-01500519 , version 1
  • OATAO : 17052

Citer

Céline Manenti, Thomas Pellegrini, Julien Pinquier. CNN-based phone segmentation experiments in a less-represented language. 17th Annual Conference of the International Speech Communication Association (INTERSPEECH 2016), Sep 2016, San Francisco, United States. pp. 3549-3553. ⟨hal-01500519⟩
135 Consultations
256 Téléchargements

Partager

Gmail Facebook X LinkedIn More