探討達成不同貼文類型適用的圖像替代文字描述方式研究

A study of the description of alt text suitable for different types of posts

近年來社群軟體的使用率提高，圖像、影像逐漸取代文字為主要的社交物件。視障者也是社群軟體的高度使用者，當視障者在使用社群軟體時，會為沒有替代文字的圖片感到困擾。臉書近年推出自動替代文字系統，該系統仍然有許多可改善空間。本研究探討另一項長期做為視障者提供視覺資訊的口述影像服務後，提出臉書貼文中的圖像替代文字應該因應貼文類型調整敘述方式。研究發現，由於相繼型的貼文在文字內容中，解釋圖像資訊的比例較少，因此貼文適合使用特定的圖像替代文字。而說明型貼文的文字內容中，已解釋許多圖像資訊，因此可以使用特定或抽象的圖像替代文字。以研究結果彙整成建議圖像替代文字產出流程圖，以提供未來相關研究或者相關服務參考。

Recently, images and videos have replaced text as the main social objects on the social software. When the visually impaired use social software, they will be troubled by pictures without alt text. Facebook has launched an automatic alt text system, which still has a lot of room for improvement. After exploring the Audio Description service that provides visual information for the visually impaired, this research proposes that the alt text of posts should be adjusted according to the types of posts. This research found that since successive posts spend a smaller proportion of the text explaining the image information, the posts are suitable for specific alt text. The text of the illustrated post contains a lot of image information, so these posts are appropriate for specific or abstract alt text. The research results are compiled into a flowchart of the different ways of producing alt text provides a reference for future related research or the development of related services.