Topic: Live subtitling of press conferences in film festivals: how easy is it?
The last few years have seen a growing demand for accessibility across language and sensory barriers. An area that has received special attention is live subtitling to make TV broadcasts and live events accessible to all. However, despite the huge progress made by speech recognition and machine translation technologies, live subtitling is not as straightforward as it may seem. Indeed, several hybrid human-machine interaction practices have developed and there is no “one size fits all” solution, as many factors (including available budget, equipment and expertise) must be taken into account. The talk will provide an overview of such hybrid practices, drawing on a recent project the author was involved in (SMART- Shaping Multilingual Access with Respeaking Technology) and will then present a practical case study focused on the challenges of subtitling the press conference announcing the films selected for the Venice Film Festival every summer. The press conference is in Italian and attracts the interest of the international press and film buffs everywhere. The event is live streamed on YouTube about a month before the beginning of the film festival and a simultaneous interpreting service (from Italian into English) is provided. By contrast, no professional subtitling is provided (in Italian or English), and the only subtitles available to viewers are the automatic ones produced via YouTube. The ARTS (Accessibility via Real Time Subtitling) project is carrying out a systematic evaluation of the accuracy that it is currently possible to achieve in this specific setting and has identified key challenges both in relation to speech recognition and machine translation. The aim of the project is to evaluate to what extent accessibility is actually ensured in this setting and draw up some best practice guidelines.
After teaching at the universities of Hull (UK), Trieste and Bologna at Forlì, in 2008 Annalisa Sandrelli joined Università degli Studi Internazionali – UNINT (Rome) as a Lecturer in English Language and Translation. She is currently the coordinator of the Interpreting section and teaches Consecutive Interpreting and Interlingual Respeaking. She has published widely on corpus-based interpreting studies, audiovisual translation, EU English, legal interpreting, and Computer Assisted Interpreter Training (CAIT). She is a member of the European Society for Translation Studies (EST), the European Association for Studies in Screen Translation (ESIST), AIA (Association of English Studies in Italy) and ESSE (European Society for the Study of English), and she is Associate member of GALMA (Galician Observatory for Media Accessibility). After participating in several EU-funded projects on legal interpreting and translation (Building Mutual Trust, Qualitas, Understanding Justice) and leading national projects on audiovisual translation (DubTalk, TVTalk, Sub!: Localisation Workflows (th)at Work 1 and 2), in recent years she has focused primarily on live subtitling. She was International Co-Investigator in the ERSC-funded SMART (Shaping Multilingual Access with Respeaking Technology 2020-2022) led by the University of Surrey, Member of the Advisory Board on the ILSA-Interlingual Live Subtitling for Access project (Erasmus+ 2017‐1‐ES01‐KA203‐037948) and Quality consultant in the LiveTextAccess (LTA) project (Erasmus+ 2018-1-DE01-KA203-004218). She is currently International Co-Investigator in SMART-UP: Shaping Multilingual Access through Respeaking Technology – Upskilling (2023-2025) and she is coordinating ARTS (Accessibility via Real Time Subtitling (2024-204). She has given talks and workshops on audiovisual translation in Italy and abroad, not only in academic conferences but also in industry events (Venice Film Festival, Turin Film Commission). As well as a trainer, she is a professional interpreter and film translator.
