Speech translation for subtitling (SubST) is the task of automatically translating speech data into well-formed subtitles by inserting subtitle breaks compliant to specific displaying guidelines. Similar to speech translation (ST), model training requires parallel data comprising audio inputs paired with their textual translations. In SubST, however, the text has to be also annotated with subtitle breaks. So far, this requirement has represented a bottleneck for system development, as confirmed by the dearth of publicly available SubST corpora. To fill this gap, we propose a method to convert existing ST corpora into SubST resources without human intervention. We build a segmenter model that automatically segments texts into proper subtitles by exploiting audio and text in a multimodal fashion, achieving high segmentation quality in zero-shot conditions. Comparative experiments with SubST systems respectively trained on manual and automatic segmentations result in similar performance, showing the effectiveness of our approach.
Dodging the Data Bottleneck: Automatic Subtitling with Automatically Segmented ST Corpora
Sara Papi;Alina Karakanta;Matteo Negri
;Marco Turchi
2022-01-01
Abstract
Speech translation for subtitling (SubST) is the task of automatically translating speech data into well-formed subtitles by inserting subtitle breaks compliant to specific displaying guidelines. Similar to speech translation (ST), model training requires parallel data comprising audio inputs paired with their textual translations. In SubST, however, the text has to be also annotated with subtitle breaks. So far, this requirement has represented a bottleneck for system development, as confirmed by the dearth of publicly available SubST corpora. To fill this gap, we propose a method to convert existing ST corpora into SubST resources without human intervention. We build a segmenter model that automatically segments texts into proper subtitles by exploiting audio and text in a multimodal fashion, achieving high segmentation quality in zero-shot conditions. Comparative experiments with SubST systems respectively trained on manual and automatic segmentations result in similar performance, showing the effectiveness of our approach.File | Dimensione | Formato | |
---|---|---|---|
2022.aacl-short.59.pdf
accesso aperto
Tipologia:
Documento in Post-print
Licenza:
Dominio pubblico
Dimensione
322.47 kB
Formato
Adobe PDF
|
322.47 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.