Long audio alignment for automatic subtitling using different phone-relatedness measures. - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Long audio alignment for automatic subtitling using different phone-relatedness measures.

Aitor Álvarez
Haritz Arzelus

Résumé

In this work, long audio alignment systems for Spanish and English are presented in an automatic subtitling scenario. Pre-recorded contents are automatically recognized at phoneme level by language-dependent phone decoders. A dynamic-programming alignment algorithm finds matches between the automatically decoded phones and the ones in the phonetic transcription for the content’s script. The accuracy of the alignment algorithm is evaluated when applying three non-binary scoring matrices based on phone confusion-pairs from each phone decoder, on phonological similarity and on human perception errors. Alignment results with the three continuous-score matrices are compared to results with a baseline binary matrix, at word and subtitle levels. The non-binary matrices achieved clearly better results. Matrix samples are given in the project’s website.
Fichier non déposé

Dates et versions

hal-01404595 , version 1 (29-11-2016)

Identifiants

  • HAL Id : hal-01404595 , version 1

Citer

Aitor Álvarez, Haritz Arzelus, Pablo Ruiz. Long audio alignment for automatic subtitling using different phone-relatedness measures. . IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP), May 2014, Florence, Italy. ⟨hal-01404595⟩
102 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More