Navigation


Hands in Hands 手口連心

【 摘 要 】 人与人的沟通交流大多以语言、表情和动作为主,而对于聋哑和特殊需要的人士来说,手语是其最不可或缺的交流方式。然而,手语本身并不通用—— 世界上有超过一亿聋哑人士,当中其98%并没有受到手语教育;与此同时,据第六次全国人口普查我国总人口数,及第二次全国残疾人抽样调查,我国的听力残疾达2054万人,语言残疾130万人,他们大多以完全失去正常沟通的能力,故手语是他们的主要沟通媒介,但由于手语并不是一门大众化和标准化的语言,就算是不同国家的手语也甚有差异,而随着计算机视觉的出现和普及,图像处理方法已经能有效的进行高准确度图像识别,在国外越来越多相关研究出现的同时,国内的手语教并不算普及,所以我们认为手语应该要受益于图像分类、检测和图像分割。

受到以上的启发,我们提出一个可以用作翻译的应用程序和学习平台 Hands in Hands(手口连心)。该应用程序旨在为提供三个服务给用户:翻译、课程和字典,以达成让普罗大众也能通过手机的翻译看懂了解手语;让聋哑人听出理解声音的目标。


AT-A-GLANCE: People use to communicate via language, facial expressions, and movements. Sign language is the most indispensable communication for the deaf and those with special needs. However, sign language is not universal - there are more than 100 million deaf people in the world, 98% of whom have not received official sign language education. What even worse is this unstandardized language varies ​​in different countries. With the emergence of computer vision, image processing methods are able to perform high-accuracy image recognition effectively. Whereas there is more related research abroad, sign language teaching is not yet considered popular in our country.
We propose an application and learning platform that can be used as a translator - Hands in Hands. The app aims to provide users with three services: translation, courses and dictionaries, so as to enable the general public to understand and understand sign language through mobile phone translation; to enable the needy to hear and understand the sound.


ROLE BRIEF: As the team leader, I mainly responsible in managing, coordinating our group. I am one of the two UIUX designers in our team, who focuses on the user behavior and reaction responses (while another UIUX designer handles the visual and drawings). Together, we design the application. Before our final outcome, we experienced through steps of user study, analysis; which we equally contributed in the early stages including interviews, brain storm, references research, analysis, storyboard etc.


TEAM

Pang Ching, Christie
Wang Siqi
Wang Yaqi
Xu Huanyu

TOOLS USED

iPad Procreate
Balsamiq Wireframes
Figma
Adobe XD, PS
Adobe AI, Animate
Adobe Premier Pro

TIMELINE

Stage 1:
Mar - Jun 2022
Stage 2:
Oct 2022 - present

MY ROLE

Usability testing
Interaction design
React to user behavior
Reaction responses test
Prototyping

Empathize and Inspiration 同理及灵感

The incompatibility of sign language education 手语教育的不普及与不通用
Communication is difficute between diasbles and general public 聋哑人士与普通人沟通存在一定困难
Lack of systematic and popular sign language education 市场上缺乏系统及普及的手语教育
Not ease to use sign language dictionary 缺乏方便易用的手语词典
Functions gap in existing software 现存为聋哑人士设计的软件没有手语与文字互译功能

Define 定义

手口连心包含翻译、字典与课程三部分,对应storyboard分别为:
We have three features in Hands in Hands. The coorresponding storyboards as follow:

Ideate 发想

我们第一步进行了用户观察,基于疫情原因,我们在项目的第二周到第三周( 3.20 - 3.27)之间观察了四位一般年轻人的习惯,观察者均为 20到 22岁的年轻大众群体,皆不是听障人士 。
我们得出下面五点结论 : 色彩运用 - 活动对象鲜明,非活动对象暗淡;前景颜色鲜明,背景颜色暗淡。限制同时显示的颜色在同一界面中不超过五种,避免将不兼容的颜色放在一起; 保持一惯性 - 包括颜色、图片、名称、字体; 需要有反馈 - 有问题时需要有提示; 快速导航 - 喜欢使用shortcut; 没有接触过听障人士或手语翻译等的听障人士或手语翻译等的app
第二步、同步进行了SWOT分析,结果如左图所示
The SWOT analysis was carried out synchronously, and the result is shown on the left.

第三步、我们采用重点采访形式,访问了八位不同背景的相关人士,其中包括四个类别:特殊学校老师、健全大众、聋哑学生、听障长者。从深度访谈中,我们整理得出了各群体的用户画像 Then, we interviewed 8 people in form of focus group. Interviewees can be categorized into: special education teachers, general public, hearing-impaired students, and elderly. From these in-depth interviews, we sorted out the user portraits of each group.

Prototype 原型

根据上述用户分析以及storyboard设计,我们梳理了工作流程图:
According to the above user analysis and storyboard design, we have sorted out the work flow chart:
基于以上提出的工作流程与功能要求,我们使用Balsamiq软件进行了低保真度模型的搭建
Based on the workflow and functional requirements proposed above, we used Balsamiq software to build a low-fidelity model.

Test 测试

我们邀请了四位测试者对我们的低保真度模型进行认知演练。针对认知演练评估结果,我们对低保真度模型进行了改进,并完善了功能,使用 Adobe XD进行了新一代模型的UI设计
We invited four interviewees to perform a cognitive test on our low-fidelity model. Based on the evaluation results of the cognitive exercise, we improved the low-fidelity model and used Adobe XD to design the UI of the new generation model

Implication and Deliverables 影响和成果