A study on constraining Connectionist Temporal Classification for temporal audio alignment - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

A study on constraining Connectionist Temporal Classification for temporal audio alignment

Yann Teytaut
  • Fonction : Auteur
  • PersonId : 1125039
Baptiste Bouvier
  • Fonction : Auteur
  • PersonId : 1185590
Axel Roebel

Résumé

Connectionist Temporal Classification (CTC) has become a standard for deep learning-based temporal alignment allowing relevant probabilistic distributions to be learned. However, by nature, CTC is a transcription objective that can be minimized without guaranteeing any alignment properties. This work aims to study several constraints to help CTC generating alignments. With a fully convolutional architecture coupled with multi-head attention, we investigate the task of phonetic alignment for clean speech and singing signals. The focus is set on the impact of additional losses, namely spectral envelope reconstruction, temporal structure invariance and guided monotony. Results show that, once scaled to have identical temporal dependence, combining all of these constraints produces best performances.
Fichier principal
Vignette du fichier
Teytaut_Bouvier_Roebel_Constraining_CTC_Alignment_Interspeech_2022 (2).pdf (1.16 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03976279 , version 1 (07-02-2023)

Identifiants

Citer

Yann Teytaut, Baptiste Bouvier, Axel Roebel. A study on constraining Connectionist Temporal Classification for temporal audio alignment. Interspeech 2022, Sep 2022, Incheon (SOUTH KOREA), South Korea. pp.5015-5019, ⟨10.21437/Interspeech.2022-10940⟩. ⟨hal-03976279⟩
68 Consultations
186 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More