Modular Audio Description: Using the extended track to enable personalisation for different audiences

模块化音频描述：使用扩展音轨为不同受众实现个性化

• 批准号: 2750384
• 金额: --
• 依托单位: University of Surrey
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2750384
• 关键词: Modular; Audio; Description; Using; extended

Video-on-demand platforms enjoyed popularity even before the current pandemic restrictions have led to increased media consumption, according to global statistics1 and the media watchdog Ofcom2. Contemporary streaming video-on-demand (SVOD) technologies continue to shape audiences' viewing practices, making media consumption increasingly adaptable and individualised. Furthermore, digital curation of user preferences has played a role in shifting genres and related scripted narratives in response to consumers' diverse tastes (Elrod 2019). Yet, consumers' diverse access needs have not been fully addressed. Streaming platforms offer access provision such as audio description (AD), i.e. a verbal commentary describing visual content, originally intended for the visually impaired. However, they do not address the demand for AD or similar assistive commentaries that meet the needs of other diverse audiences, e.g. individuals with emotion recognition difficulties (ERD). Equally important, AD offered on SVOD services is highly time constrained, as the AD narratives are delivered during gaps in the existing audio track, in line with AD convention, but often leaving little room for describing pertinent salient visual elements (Fryer 2018). However, SVOD users are no longer restricted to a particular viewing time or length of time spent consuming streamed content. This flexibility can be exploited to accommodate new practical solutions to overcome the current restricted format of AD. Therefore, the proposed research is highly timely, because it may provide foundations for an all-inclusive cost-effective AD creation workflow that meets both the demand for different AD versions for diverse user groups and the increasing production volumes.

根据全球统计数据1和媒体监管机构Ofcom2的数据，即使在当前的大流行限制导致媒体消费增加之前，视频点播平台就已经很受欢迎。当代流媒体视频点播 (SVOD) 技术继续影响观众的观看习惯，使媒体消费变得越来越适应性和个性化。此外，用户偏好的数字化管理在改变流派和相关脚本叙述以响应消费者的不同品味方面发挥了作用（Elrod 2019）。然而，消费者多样化的接入需求尚未得到充分满足。流媒体平台提供音频描述 (AD) 等访问服务，即描述视觉内容的口头评论，最初是为视障人士设计的。然而，它们并没有解决对 AD 或类似辅助评论的需求，以满足其他不同受众的需求，例如有情绪识别困难（ERD）的个体。同样重要的是，SVOD 服务上提供的 AD 时间受到高度限制，因为 AD 叙述是在现有音轨的间隙中传递的，符合 AD 惯例，但通常没有留下足够的空间来描述相关的显着视觉元素 (Fryer 2018)。然而，SVOD 用户不再受限于特定的观看时间或消费流媒体内容所花费的时间长度。可以利用这种灵活性来适应新的实用解决方案，以克服当前 AD 的受限格式。因此，所提出的研究是非常及时的，因为它可以为全面的、具有成本效益的AD创建工作流程提供基础，满足不同用户群体对不同AD版本的需求和不断增加的生产量。