Podcasts are becoming an increasingly popular way to share streaming audio content. Podcast summarization aims at improving the accessibility of podcast content by automatically generating a concise summary consisting of text/audio extracts. Existing approaches either extract short audio snippets by means of speech summarization techniques or produce abstractive summaries of the speech transcription disregarding the podcast audio. To leverage the multimodal information hidden in podcast episodes we propose an end-to-end architecture for extractive summarization that encodes both acoustic and textual contents. It learns how to attend relevant multimodal features using an ad hoc, deep feature fusion network. The experimental results achieved on a real benchmark dataset show the benefits of integrating audio encodings into the extractive summarization process. The quality of the generated summaries is superior to those achieved by existing extractive methods.

Leveraging multimodal content for podcast summarization / Vaiani, Lorenzo; LA QUATRA, Moreno; Cagliero, Luca; Garza, Paolo. - ELETTRONICO. - (2022), pp. 863-870. ((Intervento presentato al convegno ACM/SIGAPP Symposium on Applied Computing tenutosi a Virtual, Online nel April 25th 2022 - April 29th 2022 [10.1145/3477314.3507106].

Leveraging multimodal content for podcast summarization

Lorenzo Vaiani;Moreno La Quatra;Luca Cagliero;Paolo Garza
2022

Abstract

Podcasts are becoming an increasingly popular way to share streaming audio content. Podcast summarization aims at improving the accessibility of podcast content by automatically generating a concise summary consisting of text/audio extracts. Existing approaches either extract short audio snippets by means of speech summarization techniques or produce abstractive summaries of the speech transcription disregarding the podcast audio. To leverage the multimodal information hidden in podcast episodes we propose an end-to-end architecture for extractive summarization that encodes both acoustic and textual contents. It learns how to attend relevant multimodal features using an ad hoc, deep feature fusion network. The experimental results achieved on a real benchmark dataset show the benefits of integrating audio encodings into the extractive summarization process. The quality of the generated summaries is superior to those achieved by existing extractive methods.
File in questo prodotto:
File Dimensione Formato  
ACM_SAC2022_MATeR_preprint.pdf

accesso aperto

Descrizione: Post-print version without editor layout
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 2.98 MB
Formato Adobe PDF
2.98 MB Adobe PDF Visualizza/Apri
3477314.3507106.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 3.18 MB
Formato Adobe PDF
3.18 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11583/2963408