Speech and translation technologies for voice-over and audio description: final results of the ALST project
Matamala, Anna (Universitat Autònoma de Barcelona)

Date: 2015
Abstract: The ALST project (FFI-201231024, funded by the Spanish Ministry of Economy) started in January 2013 aiming to research the implementation of speech technologies (speech recognition and speech synthesis) and translation technologies (machine translation) in two audiovisual transfer modes: audio description, and voice-over. The project will end in December 2015, and the presentation will give an overview of the main objectives of the project as well as its main results. Although limited in its funding and scope, the project has carried out innovative research in the following aspects: a) Implementation of automatic speech recognition and respeaking to generate faster transcripts: experiments have been carried out with professional transcribers to compare various working scenarios (manual transcription/automatic speech recognition/respeaking). b) Implementation of machine translation plus post-editing as an alternative to traditional audiovisual translation: experiments have been carried out using audio descriptions (English into Catalan) and also wildlife documentaries to be voiced over (English into Spanish). c) Implementation of text-to-speech systems instead of human voices, both in audio descriptions and documentaries. All experiments have in common the willingness to obtain objective data (generally linked to the productivity gain in terms of time) but also subjective data on user experience and perceived quality. The presentation will not get into a detailed analysis of all experiments for obvious time constraints, but will offer a general structured overview that will allow the audience to understand the main results of the project as well as possible future research venues.
Note: Número d'acord de subvenció FFI2012-31024, 2014SGR0027
Rights: Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, sempre que no sigui amb finalitats comercials, i sempre que es reconegui l'autoria de l'obra original. Creative Commons
Language: Anglès.
Document: conferenceObject
Subject: Traducció audiovisual ; Audiodescripció ; Veus superposades ; Tecnologies ; Audiovisual translation ; Audio description ; Voice-over ; Technology ; Traducción audiovisual ; Audiodescripción ; Voces superpuestas ; Tecnología
Published in: 6th International Conference Media for All. Sydney, Austràlia, 6a : 2015

15 p, 101.9 KB

The record appears in these collections:
Research literature > UAB research groups literature > Research Centres and Groups (scientific output) > Arts and Humanities > TransMedia Catalonia
Contributions to meetings and congresses > Presentations

 Record created 2015-09-21, last modified 2018-10-28

   Favorit i Compartir