Files

42061.pdf (5.85 MB)
42061.mp3 (25.5 MB)

Date

2017-03-04

Authors

Mettouchi, Amina

Bettinson, Mat

Bird, Steven

Contributor

Advisor

Department

Instructor

Depositor

Speaker

Mettouchi, Amina
Bettinson, Mat
Bird, Steven

Researcher

Consultant

Interviewer

Annotator

Journal Title

Journal ISSN

Volume Title

Publisher

Volume

Number/Issue

Starting Page

Ending Page

Alternative Title

Abstract

Description

The oral transmission of recipes provides a rich context for collection of procedural discourse. In the diaspora, cuisine may be maintained longer than language, and the transmission of recipes to another generation offers an occasion for reconnection with older speakers. While it is easy to video the preparation of a dish, the resulting video is harder to navigate and harder to translate, compared with the approach we have developed. We have developed a kind of digital storytelling app(1) Zahwa which captures a recipe as a sequence of still images aligned to an audio narrative. The user photographs ingredients and utensils, then photographs each stage of preparation. The images are opened inside the app, and user presses a record button and explains the recipe, swiping to advance to the next image. Finger gestures are captured, so that reference to items within an image are unambiguous. During playback of a recipe, a user hears the audio and watches a slideshow of the images, and sees image highlights. The user navigates the recipe by swiping to get to the desired image and then touches it to resume playback. Recipes can be orally translated, so that users can experience the slideshow aligned with multiple languages. The collection of recordings obtained in this way forms a spoken corpus. Common elements, such as verbally annotated images of ingredients and utensils, offer the chance to collect dialect variants, and link them to a domain-specific lexicon. Syntactic constructions associated with procedural discourse can be matched with their equivalents in other languages. The corpus also offers contextualized teaching materials, providing learners with time-aligned images. The method has been implemented as a web app, supporting the usual social media features of sharing and liking. Recipes are stored in the cloud, when bandwidth is available, and shared via URL. The images and audio can be initially captured in high resolution, and shared over wireless networks in lower resolution until a wifi connection is available when the high resolution version is secured. A web API provides programmatic access to the data. The approach offers an oblique method for linguists to collect language documentation, in which speakers of an endangered language may be independently motivated to describe recipes, crafts, works of art, etc, in short, anything that can be adequately represented as a series of images with an accompanying audio track. We report results of using this method in schools and in conventional language documentation projects. (1) cf. Bidwell, N, T Reitmaier, G Marsden, S Hansen (2010). Designing with mobile digital storytelling in rural Africa, Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp 1593-1602, Association for Computing Machinery

Keywords

Citation

URI

http://hdl.handle.net/10125/42061

Extent

Format

Geographic Location

Time Period

Related To

Table of Contents

Rights

Rights Holder

Local Contexts

Collections

5th International Conference on Language Documentation and Conservation (ICLDC)