More haste less speed: edited versus verbatim respoken subtitles
More haste less speed: edited versus verbatim respoken subtitles
复制标题
欲速则不达：编辑字幕与逐字字幕
DOI：
--
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
Pablo Romero Fresco
中科院分区：
文献类型：
--
作者：
Pablo Romero Fresco
在为聋人和听力障碍者 (SDH) 制作字幕时，编辑字幕和逐字字幕之间的选择一直是一个有争议的问题。尽管学者们通常支持编辑，但聋人协会却倾向于要求逐字字幕，作为充分观看视听节目的唯一途径。现在，欧洲立法使 SDH 不再是一种特权，而是所有观众的一项权利，这种对逐字字幕的需求也已扩展到直播节目。然而，现在的实时字幕主要由语音识别（复述）产生，呈现出不同的情况并需要不同的分析。本文的目的是提供对口述字幕的描述，特别是关于其速度的描述。首先，概述了涉及字幕速度问题的不同各方，然后回顾了迄今为止进行的研究以及由此实施的指南，特别关注英国。然后，对 BBC 播放的十个重播节目进行了分析，提供了有关原始配乐速度、重播字幕速度、编辑量以及在此过程中丢失的信息的数据。该分析获得的结果表明，至少在所分析的节目中，很少制作逐字字幕。有人认为，目前由复述者进行的编辑造成的信息损失最小，特别是与观众以当前速度阅读复述字幕可能造成的信息损失相比。
The choice between edited and verbatim subtitles has always been a controversial issue in subtitling for the deaf and hard-of-hearing (SDH). Whereas scholars often support editing, deaf associations tend to demand verbatim subtitles as the only way to have full access to audiovisual programmes. Now that European legislation is making SDH no longer a privilege but a right for all viewers, this demand for verbatim subtitles has also been extended to live programmes. Yet, live subtitles, nowadays mostly produced by speech recognition (respeaking), present a different situation and require a different analysis. The aim of this article is to provide a description of respoken subtitles, especially with regard to their speed. First of all, an overview is given of the different parties involved in the issue of subtitling speed, followed by a review of the research carried out so far and of the guidelines that have been implemented as a result, with particular focus on the UK. Then, an analysis is presented of ten respoken programmes broadcast by the BBC, providing data regarding the speed of the original soundtrack, the speed of respoken subtitles, the amount of editing carried out and the information lost in this process. The results obtained in this analysis show that verbatim respoken subtitles, at least in the programmes analysed, are rarely produced. It is argued that editing, as currently carried out by respeakers, causes a minimal loss of information, especially as compared to the potential loss of information for viewers reading respoken subtitles at the current speeds.