<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Arun George on Medium]]></title>
        <description><![CDATA[Stories by Arun George on Medium]]></description>
        <link>https://medium.com/@arungeorge_?source=rss-1737a51393c7------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*d8s47Q7qnG7zTiSEziMG0w.jpeg</url>
            <title>Stories by Arun George on Medium</title>
            <link>https://medium.com/@arungeorge_?source=rss-1737a51393c7------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 20:36:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@arungeorge_/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[New portfolio page.]]></title>
            <link>https://medium.com/@arungeorge_/new-portfolio-page-545ec01ef722?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/545ec01ef722</guid>
            <category><![CDATA[ux-portfolio]]></category>
            <category><![CDATA[portfolio]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Thu, 25 Dec 2025 21:25:43 GMT</pubDate>
            <atom:updated>2025-12-25T21:25:43.615Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://www.arungeorge.me/">Arun George | AI Conversation Design Architect</a></p><p>Will be adding case studies.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=545ec01ef722" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Made Simple -What Every Conversation Designer Should Know (Series) — RAG Basics]]></title>
            <link>https://uxplanet.org/ai-made-simple-what-every-conversation-designer-should-know-series-rag-basics-b360ffd667f7?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/b360ffd667f7</guid>
            <category><![CDATA[conversational-ai]]></category>
            <category><![CDATA[chatbot-design]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[retrieval-augmented-gen]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Mon, 12 May 2025 21:37:56 GMT</pubDate>
            <atom:updated>2025-05-15T15:24:18.502Z</atom:updated>
            <content:encoded><![CDATA[<h3>AI Made Simple -What Every Conversation Designer Should Know (Series) — RAG Basics</h3><p>As a conversation designer, it’s important to understand some of the techniques used to optimize large language models (LLMs). While you may not be implementing these methods all by yourself, having a working knowledge of them can definitely help you collaborate more effectively with engineers, researchers, or speech scientists — and ultimately make you a better designer.</p><p>Let’s start with <strong>RAG</strong>.</p><p><strong>So, what exactly is RAG?</strong></p><p>RAG stands for <strong>R</strong>etrieval-<strong>A</strong>ugmented <strong>G</strong>eneration. It’s a technique used to make LLMs smarter and more accurate by combining their natural language abilities with your own external data sources or knowledge base.</p><p><strong>Let’s imagine a simple scenario</strong>:<br>You’re designing an ai-powered chatbot for a company with frequently updated FAQs or documentation. Even if you’re using a powerful model like ChatGPT or Claude, there’s a challenge you’d encounter— these models don’t know about your updates. Why? because these large language models are trained on static datasets and re-training them every time your content changes can be impractical and expensive.</p><p><strong>This is where RAG comes in.</strong></p><p>Instead of relying solely on what the model “knows,” RAG retrieves relevant documents from your knowledge base (like an FAQ or internal wiki) in real time. These documents are then passed into the prompt as context — so the model can generate a response that’s both accurate and grounded in your latest data.</p><blockquote>In simple terms:<br><strong> Retrieve relevant info from your data<br> Augment the language model with that info<br> Generate the final response</strong></blockquote><p><strong>Let’s visualize the process.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*NQ-GmTGotV-Yk3pbPLJx7g.png" /><figcaption>RAG Generation workflow (Source: bentoml images)</figcaption></figure><p><strong>What’s Happening Behind the Scenes?</strong><br>Before your data can be used by the language model to generate a response, it first needs to be processed and transformed. Here’s how that typically works:</p><p><strong>Chunking the Content</strong><br>Whether you’re using FAQs, documents, website content, or even images, the first step is to break them into smaller chunks. For text, this could be paragraphs or sections. For images or audio, these may be frames or short clips.</p><p><strong>Creating Embeddings</strong><br>Now each chunk is then converted into an embedding — a list of numbers that represents the meaning or features of that content. For text, these embeddings capture the <em>semantic meaning</em>. For images, models like OpenAI’s CLIP convert visual content into vector form. For audio, embeddings represent sounds or spoken words.</p><p><strong>Storing in a Vector Database</strong><br>These numerical representations — also known as vectors — are stored in a vector database, which is optimized to search for similar content. When a user asks a question, the system retrieves the most relevant chunks based on semantic similarity — not just keywords.</p><p><strong>Retrieval and Augmentation</strong><br>The most relevant chunks are retrieved from the vector database and added to the language model’s prompt. This gives the model grounded context to generate a response that’s accurate, up-to-date, and specific to your data.</p><p><strong>Below is an example illustrating what happens when a user initiates a search in a RAG‐enabled application.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/973/1*NKFPj35C-dL-TYD9Qa2xuA.png" /><figcaption>The user’s question is embedded and searched against a vector database, relevant document chunks are retrieved to construct an augmented prompt, and then sent to the LLM to produce the final answer.</figcaption></figure><blockquote>User Query<br> ↓<br>Embed Query<br> ↓<br>Search Vector DB<br> ↓<br>Retrieve Relevant Chunks<br> ↓<br>Construct Augmented Prompt (using query + retrieved chunks)<br> ↓<br>Pass/Send to Language Model<br> ↓<br>Generated Answer</blockquote><p>As conversation designers, understanding RAG (Retrieval Augmented Generation) unlocks new creative possibilities in how we structure content and support our assistants.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b360ffd667f7" width="1" height="1" alt=""><hr><p><a href="https://uxplanet.org/ai-made-simple-what-every-conversation-designer-should-know-series-rag-basics-b360ffd667f7">AI Made Simple -What Every Conversation Designer Should Know (Series) — RAG Basics</a> was originally published in <a href="https://uxplanet.org/">UX Planet</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Voice UX Process Under 5 Mins]]></title>
            <link>https://uxplanet.org/voice-ux-process-81812c8ccf70?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/81812c8ccf70</guid>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[alexa]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[voice-user-interfaces]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Sun, 11 Feb 2018 07:36:10 GMT</pubDate>
            <atom:updated>2022-02-06T16:34:48.052Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ec9iLRqXiCzl775SJBSUlw.png" /></figure><p>It seems there’s always something to be mentioned about voice interfaces these days. Now that’s a great thing, really. Voice after all is the only interface that doesn’t require a training manual…unless you design horribly. If you do, we’ll be repeating the same horrible experience <a href="https://www.wired.com/2012/10/1-annoying-technologies-as-chosen-by-the-wired-commenters/">we have on automated phone systems</a>. To avoid this, here are some things one should think about when designing for voice. It’s a process that I have used with a team of other Voice UX designers back when I was working as a consultant to Fortune 500 companies. I hope you’d find some value. If you want to learn more about this, you can get my course on this <a href="https://www.udemy.com/designing-voice-user-experience/">here</a>.</p><p><strong>Context Gathering<br></strong>Before we design anything, we need to discover some things. Actually, a lot of things. Context gathering will help us do just that. In fact, it’s the core of user-centered design process. Here we conduct stakeholder interviews to understand business goals and aims. We look at current experience and how the users are solving the problem, competitive analysis to see how we can stand apart. We also look at the frequency of usage to find out why most people use your system and finally brand persona. Your brand persona is a very important aspect of your voice design and if you want to read more on understanding persona, I’ve written an article you should <a href="http://uxdesign.cc/voice-user-experience-design-and-prototyping-for-mere-mortals-ef080c843640">check it out.</a></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FRZ_OYUT6ls0%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DRZ_OYUT6ls0&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FRZ_OYUT6ls0%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/1002b22f3f7ebc8e3055db263f98fb75/href">https://medium.com/media/1002b22f3f7ebc8e3055db263f98fb75/href</a></iframe><p><strong>High-Level Design<br></strong>Once we’re done with context gathering, we can move on to high-level design phase. Here we look at summary of our context gathering to understand the speech system better. First we create a high-level information architecture or call-flows to look at the speech system we’ll be designing in regards to logic. Remember that speech happens in a context and not call-flows . So when creating high-level flows, make sure to keep it ‘high level’ and not write dialog prompts on top of it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*1UPfE7RJvbliUVsQSNWL_g.png" /></figure><p>Then we can move on to writing some sample dialogs in persona. What’s a sample dialog? Simply put — A sample dialog is a script that shows a conversational flow between the system and the user. You should create as much as you feel is needed but focus on most of the core tasks based on frequency of usage. Build out both happy paths (where things go accordingly) and failure paths to see how you could design error strategies. Sample dialogs can also be helpful to write conversational prompts to avoid technical jargons or unnecessary words.</p><p><strong>Prototype &amp; Research <br></strong>A design that’s not based on data or testing is simply an assumption which could cost you time and money. When it comes to prototyping and running usability testing for voice, we often use <a href="https://en.wikipedia.org/wiki/Wizard_of_Oz_experiment">Wizard of WOz method</a>, commonly known as WOz. So let’s identify key tasks to test and build out some scenarios. Then have each participant use the system as they’d in real-life. Then get their feedback by probing without leading them. Look for task success and CSAT (Customer Satisfaction Score). Then analyze results to see if you need to run additional iteration or not. To learn more about CSAT, <a href="https://blog.emolytics.com/customer-experience/customer-satisfaction-score-csat-kpi/">check it out here</a>. Or if you want to find out how many participants you should test, here’s <a href="https://www.nngroup.com/articles/why-you-only-need-to-test-with-5-users/">someone</a> who can explain better.</p><p><strong>Detailed Design &amp; Deploy<br></strong>Well, this is exciting! Now we can finally move on to some actual design. A detailed design is….well, very detailed. They can be considered a <em>state-by-state </em>documentation on how your VUI (Voice User Interface) should behave — including the initial prompt, no-match errors in case your users say something out of context, conditions to follow next and so on. I’ve written more on that <a href="http://justinmind.com/blog/voice-user-experience-design-and-prototyping-for-mere-mortals/">here</a>. This is also where you want to start thinking about implementing <a href="https://developer.amazon.com/docs/custom-skills/speech-synthesis-markup-language-ssml-reference.html">SSML</a> (Speech Synthetic Mark-up Language), so your actual design would sound natural and conversational.</p><p>It’s important to involve the dev team for a VUI review to make sure your design can be implemented. The dev team might have some changes or you might have to make some changes to the dialog structure as a result. Whatever it is, be on the same page and then you can deliver the final VUI for implementation.</p><p>While we’re done with the actual design, it’s important to monitor the VUI in production environment for continuous improvement. Seeing the trend of voice-enabled devices and apps coming out at rapid speed, my fear is that without actual understanding or having a good process in place, we’d start hearing terrible speech applications and start hating these experiences the way we often do when it comes to <a href="https://www.comparebusinessproducts.com/contact-center/why-customers-hate-ivr-systems">automated speech systems</a>. When done properly, voice-enabled interfaces can be extremely empowering and helpful. After all, we’re <a href="https://books.google.com/books/about/Wired_for_Speech.html?id=45_bAAAAMAAJ">wired for speech</a>. I hope this article is helpful and let’s start creating some world-class voice experiences!</p><p><strong>More articles from me</strong>:</p><p>- <a href="https://uxdesign.cc/voice-user-experience-design-and-prototyping-for-mere-mortals-ef080c843640">Designing Voice Experience</a><br>- <a href="https://www.youtube.com/watch?v=RZ_OYUT6ls0">Building blocks of AI Chatbots</a></p><p><strong>Get in touch</strong><br>- <a href="https://twitter.com/arungeorge_">https://twitter.com/arungeorge_</a><br>-<a href="https://www.linkedin.com/in/linkarungeorge/">https://www.linkedin.com/in/linkarungeorge/</a></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Fb3e808%3Fas_embed%3Dtrue&amp;dntp=1&amp;display_name=Upscribe&amp;url=https%3A%2F%2Fupscri.be%2Fb3e808%2F&amp;image=http%3A%2F%2Fapi.screenshotlayer.com%2Fapi%2Fcapture%3Faccess_key%3Dfe59908dad3baab69ffab249a2224b03%26viewport%3D1024x612%26width%3D1000%26url%3Dhttps%253A%252F%252Fupscri.be%252Fb3e808%253Fscreenshot&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/96d08ab34921bdd17986cb5c0396842f/href">https://medium.com/media/96d08ab34921bdd17986cb5c0396842f/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=81812c8ccf70" width="1" height="1" alt=""><hr><p><a href="https://uxplanet.org/voice-ux-process-81812c8ccf70">Voice UX Process Under 5 Mins</a> was originally published in <a href="https://uxplanet.org/">UX Planet</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[UX Process for voice]]></title>
            <link>https://medium.com/@arungeorge_/ux-process-for-voice-2f25c2ad11dc?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/2f25c2ad11dc</guid>
            <category><![CDATA[voice-user-interfaces]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[ux-process]]></category>
            <category><![CDATA[user-experience]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Sat, 03 Feb 2018 19:23:53 GMT</pubDate>
            <atom:updated>2018-02-11T05:00:38.908Z</atom:updated>
            <content:encoded><![CDATA[<p>Context Gathering<br> — Business goals/stakeholder interviews<br> — Core frequent usage of tasks<br> — Competitive analysis<br> — Understanding brand’s persona</p><p>High-Level Design</p><p>— Building information architecture <br> — Understanding persona<br> — Conversational design <br> — Writing sample dialogs</p><p>Prototype &amp; Research</p><p>— Identify tasks<br> — Wizard of Oz usability testing<br> — Analyzing results</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2f25c2ad11dc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing Voice Experience]]></title>
            <link>https://uxdesign.cc/voice-user-experience-design-and-prototyping-for-mere-mortals-ef080c843640?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/ef080c843640</guid>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[bots]]></category>
            <category><![CDATA[voice-user-interfaces]]></category>
            <category><![CDATA[alexa]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Thu, 28 Sep 2017 13:28:11 GMT</pubDate>
            <atom:updated>2025-05-21T22:51:39.247Z</atom:updated>
            <content:encoded><![CDATA[<h4>As a voice user experience designer, I’ve worked with Fortune 500 companies to help improve their speech applications. In this article, I want to share my experience on what are some of the things you should think about when designing for voice.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ySnj3RRtvBZEuU0y." /><figcaption>(Google Images)</figcaption></figure><p>The rise of voice interfaces is pretty obvious as every single major company rolls out their idea of a virtual assistant. With voice interfaces, all of a sudden we can search, send messages, even control our connected devices seamlessly — all by our natural voice. But there’s a catch — according to VoiceLabs, <a href="https://www.recode.net/2017/1/23/14340966/voicelabs-report-alexa-google-assistant-echo-apps-discovery-problem"><em>69 percent of the 7,000-plus Alexa “Skills” — voice apps, if you will — have zero or one customer review, signaling low usage.</em></a> Now that’s really not good statistics. So why is that? I believe <a href="https://venturebeat.com/2017/07/16/the-rise-of-the-voice-interface/">VentureBeat</a> has said it best:</p><blockquote>“ We are not creating conversations, we are building old-school commands hidden behind voice requests.”</blockquote><p>With these voice interfaces becoming more and more common, the need to make them more conversational and user-centered is significant. In this article, I’ll explain my process for voice design based on my experience of working with clients from variety of industries from telecommunications to retails.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FRZ_OYUT6ls0%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DRZ_OYUT6ls0&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FRZ_OYUT6ls0%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/1002b22f3f7ebc8e3055db263f98fb75/href">https://medium.com/media/1002b22f3f7ebc8e3055db263f98fb75/href</a></iframe><h3>What is a Voice User Interface?</h3><p>Let’s visit the concept of what exactly a voice user interface is. A Voice User Interface, or VUI (pronounced “voo-e”), is an interface that the user interacts with by communicating vocally. Most of us are familiar with voice interfaces by interacting with automated phone systems. Sadly, a lot of phone systems have a very badly designed interface. There’s a reason behind this, and it’s because the developers who design these systems don’t understand <strong>how to design for voice experience</strong>.</p><p>Some of the reasons why you have such bad experience with voice devices:</p><ul><li>They lack context in speech and not truly conversational in nature.</li><li>They’re designed to act as, “information collectors.”</li><li>The dialogues are spoken the way we write and not speak.</li></ul><p>It’s important to realize that it’s usually not the technology that’s causing a bad experience but rather the design interfaces you’re interacting with. Let’s look at some ways we can get started on designing a great experience for your voice devices.</p><h3><strong>Understanding Persona</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/0*UPJho7dbJqPc8BnP." /><figcaption>(Source: pexels.com)</figcaption></figure><p>Everything has a personality, even your bots and voice devices. I remember a friend who once told me that he prefers Amazon Alexa over Google Home since he can wake Alexa up by using a name rather than <em>OK Google</em>. To my friend, the name and the voice made it more human. He was referring to its <a href="https://en.wikipedia.org/wiki/Persona_(user_experience)">persona</a> and it is relevant as technology tries its best to connect with its users.</p><p>Clifford Nass, who was a professor of Communications at Stanford University, and a renowned authority on human-computer interaction claimed in his book, <a href="https://www.amazon.com/gp/product/1575860538/ref=as_li_tl?ie=UTF8&amp;camp=1789&amp;creative=9325&amp;creativeASIN=1575860538&amp;linkCode=as2&amp;tag=theuxfactor-20&amp;linkId=cbeabbb8ca8a88961be03c4517610dd7"><em>The Media Equation</em></a><em>:</em></p><blockquote>People tend to treat computers and other media as if they were either real people or real places. (The Media Equation)</blockquote><p>We’re emotional beings and tend to bring that association to everything. Don Norman, a giant in the field of Human-Computer- Interaction, even wrote a book on this called <a href="https://www.amazon.com/Emotional-Design-Love-Everyday-Things/dp/0465051367/ref=as_sl_pc_qf_sp_asin_til?tag=theuxfactor-20&amp;linkCode=w00&amp;linkId=ace39e4511bafadd994a01c6dc7ee6da&amp;creativeASIN=0465051367"><em>Emotional Design</em></a>. Norman states:</p><blockquote>People can more easily relate to a product, a service, a system, or an<br>experience when they are able to connect with it at a personal level. (Emotional Design)</blockquote><p>Wanna break into conversation design? Check out my YouTube channel for more info: <a href="https://youtu.be/6_86LIrcvOo">https://youtu.be/6_86LIrcvOo</a></p><p>As you can tell, great user experience is not about just making usable or functional apps but also about creating an emotion. Likewise, a great persona in voice is not about just having a pretty voice. It’s also about connecting with the user on the other end. When we hear a voice, we unconsciously make a lot of assumptions about that person. These assumptions include how intelligent that person might be or which region or country they’re from.</p><p>A persona or personality can be thought of as a <em>character</em> in the voice user interface world, just the same way in a film or book. It’s one way a company can brand itself, for it’s an extension of the brand itself. Therefore, it’s very important to pick the persona carefully. I remember running a usability study with a major healthcare brand and hearing comments from participants stating, “<em>That voice sounds too happy. I’m calling to refill a prescription. Why does she sound too happy?” </em>We eventually had to coach the voice talent to adjust the prompts to match the proper tone. One common misunderstanding is you can hire a voice actor with a pleasant voice and that’d be the end of persona creation.</p><p>Creating a well-crafted persona takes time and research. According to the book, <a href="https://www.amazon.com/gp/product/0321185765/ref=as_li_tl?ie=UTF8&amp;camp=1789&amp;creative=9325&amp;creativeASIN=0321185765&amp;linkCode=as2&amp;tag=theuxfactor-20&amp;linkId=f5cec037e18185a783c751cf57f4143f"><em>Voice User Interface Design</em></a>, here are some of the things one should think about when designing a persona for your voice interface.</p><ul><li>The Role: What’s the role of the application to the user? In other words, is it an assistant that the user is familiar with that gives advice on stock options? Is it a bank clerk?</li><li>Company Brand / Image: The persona that you pick for the system or application should be at least compatible with the brand or company’s image.</li><li>Familiarity &amp; Target Audience: Your persona should be familiar to your users. Therefore, for a compelling persona, we need to consider demographics, attitudes, frequency of usage, and the lifestyle of the user. A persona that works well in one culture might not work in another.</li></ul><h3><strong>Designing the Blueprint</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/0*nbGovTyRvHRfCxvl." /><figcaption>(Source: pexels.com)</figcaption></figure><p>While it’s very tempting to design your voice experience using a flow-chart, it’s not the best way to achieve a great experience. Although it’s important to have a flow-chart explaining the workflow or information architecture (see below), the conversational design prompts shouldn’t be focused on that. Remember that speech happens in a context and is not strictly based on logic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*1UPfE7RJvbliUVsQSNWL_g.png" /></figure><p>(source: Voice User Interface Design book)</p><p>How many automated systems have you heard with prompts such as, ‘If you want to return to the main menu, press 1’. This is not recommended since we tend to move conversations <em>forward</em> and not <em>backward</em>. For example, if I ask you to repeat something, you don’t go back and retrieve it. Instead, you simply state differently — i.e., “Sure, here’s it again”. The more I ask you to repeat, the more differently you explain. The ingredient to make a design naturally conversational is understanding <em>linguistics</em>. In a conversational design interface, the focus is on building an interaction based on the way people <em>speak</em>, not how they <em>write</em>. A lot of voice interfaces are written in the way we write rather than the way we speak.</p><h3><strong>Designing a VUI dialog</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*0p5a5dL0lXDzv11u.jpg" /><figcaption>(Source: allexpress.com)</figcaption></figure><p>In voice interfaces, we don’t have wireframes; we have sample dialogues. But before we get to that, let’s start with user stories. Start by describing scenarios in which your users will find useful. Identify major scenarios to describe the purpose of your interface and ways your users can interact with them.</p><p>Next, start crafting a few sample dialogs for your scenarios. <em>A sample dialog is a script that shows a conversational flow between the system and the user</em>. You can think of this as the VUI version of a wireframing process. A sample dialog can be something you show your client to get feedback before you start prototyping.</p><h3><strong>A Sample Dialog</strong></h3><p>Here’s an example of what a sample dialogue might look like:</p><h4>Caller using the phone to make a payment; payment info stored.</h4><p><strong>System:</strong> Thanks for calling BankOMatic. I looked up your number and found an account. I see you have a payment that’s due today. Are you calling about that?</p><p><strong>Caller:</strong> Yes.</p><p><strong>System:</strong> OK. For security, tell me the last four digits of your Social Security number.</p><p><strong>Caller:</strong> 4321.</p><p><strong>System:</strong> Thanks, let me look it up. <br>Found it. You have a payment of $40.98 that’s due today. <br>Would you like to make a payment now?</p><p><strong>Caller: </strong>Yep</p><p><strong>System:</strong> All right, I see you have a VISA card ending in 7894 on file; would you like me to use that?</p><p><strong>Caller:</strong> Yes, please.</p><p><strong>System:</strong> Just a second… All done! I’ve put that through. Just so you know, it might take up to 48 hours to show up on your account. Now, if that’s all you needed, feel free to hang up, and thanks for using BankOMatic.</p><p>While the dialog above is a happy journey, you should also write for not-so-happy scenarios such as, what happens if the user says something out-of-context, error progression, etc. We’ll talk more about errors later on. You probably notice the use of contraction in the dialogue above. In a conversation, we tend to use contractions (i.e. ‘You’re,’ ‘I’m’). And it’s not because we’re lazy. There are some things that would sound really odd if we don’t use a contraction. For example, how would you say the following without the use of contraction?</p><ul><li>“Wasn’t that presentation great?” → “Was not that presentation great?”</li></ul><p>The question above makes you want to say, “Yes, your honor. It was indeed a great one!”</p><p>Whilst we do this unconsciously, it’s important to realize once again — the way we speak is different from the way we write, so please do keep that in mind when designing your VUIs.</p><p>Another idea to keep in mind is what’s called a <a href="http://dictionary.cambridge.org/us/grammar/british-grammar/discourse-markers/discourse-markers-so-right-okay"><strong>discourse marker</strong></a>. A discourse marker is a word or phrase that you can use to connect and arrange what you say or write (i.e. ‘anyway,’ ‘now’, etc.). In the dialogue above, I have used the word ‘Now’ to make an obvious transition from one idea to another.</p><p>You see a lot of speech systems these days using technical jargon or words that are often only familiar to engineers and developers. Let’s take this example:</p><p><strong>System:</strong> Your request has been processed.</p><p>How would you rewrite in everyday language? Here’s one way:</p><p><strong>System:</strong> Done! You’re all set!</p><h3><strong>Error Strategies</strong></h3><p>In early 2017, Google published <a href="https://developers.google.com/actions/design/conversation-repair"><em>In Conversation, There Are No Errors</em></a><em> , </em>and according to the article,<em> </em>one should think of errors as opportunities to create meaningful conversations. I’ll let you read the article on the principles behind handling errors on your own time. For now, I’ll just jump into some examples.</p><h4>Rapid re-prompts</h4><p>The rapid re-prompt approach doesn’t provide detailed information right away. This is similar to the kinds of statements people might use in a typical conversation to show that they didn’t really understand the speaker:</p><ul><li>“What was that?”</li><li>“Say it again?”</li><li>“I’m sorry?”</li></ul><h4>Escalating Errors</h4><p>From my personal experience of running usability studies with clients from variety of industries, I’ve found that escalating errors can save time and task completion; especially for power users.</p><p><strong>System: </strong>What’s your date of birth?<br><strong>Caller:</strong> Uh…<br><strong>System:</strong> Just tell me your date of birth using 2 digits for the month, 2 digits for the day, and 4 for the year.</p><p>As you can see, the error strategy started escalating from general to specific rather than just giving all of the information right away. It’s also great for power users who are used to hearing the prompts so many times.</p><h4>Be Cooperative</h4><p>When writing prompts and error strategies, consider being cooperative by applying <a href="http://www.lancaster.ac.uk/fass/projects/stylistics/topic12/14cp1.htm">Grice’s Maxim</a>s. While Grice’s Maxims offer a useful framework, they’re only a starting point — language is inherently complex and socially contextual. Our design choices should be guided by how users actually behave in real-world interactions. Not only this will help the user experience, but would also create confidence and trust in your users when interacting with the device.</p><ul><li><strong>The maxim of quantity —</strong> try to<strong> </strong>be as informative as one possibly can, and give as much information as is needed, and no more.</li><li><strong>The maxim of quality — </strong>try to<strong> </strong>be truthful, and don’t give information that’s false or not supported by evidence.</li><li><strong>The maxim of relation —</strong> try<strong> </strong>to be relevant, and say things that are pertinent to the discussion.</li><li><strong>The maxim of manner —</strong> try to be as clear; avoid obscurity and ambiguity.</li></ul><h3><strong>Final Thoughts</strong></h3><p>While there are many ways you can design a great experience for voice, remember that the end focus is on users like you and I. So, it’s important to understand not only the target users but also the context in which the dialogues will appear.</p><p>In other words, study the way we speak and write. Study <a href="https://en.wikipedia.org/wiki/User-centered_design"><strong>user-centered design processes</strong></a><strong> </strong>and learn how to approach the challenge humanly. It’s usually not the limitation of speech technology that’s responsible for a horrible voice experience. It’s usually the designers not knowing how to apply these processes that result in a less-than-desirable voice interface. Hopefully, this article will help you design for mere mortals.</p><h3>Resources</h3><ul><li><a href="https://www.youtube.com/watch?v=RZ_OYUT6ls0">Building Blocks of AI Chatbots</a></li><li><a href="https://www.amazon.com/gp/product/0321185765/ref=as_li_tl?ie=UTF8&amp;camp=1789&amp;creative=9325&amp;creativeASIN=0321185765&amp;linkCode=as2&amp;tag=theuxfactor-20&amp;linkId=598b72c41c27359a829bb565ab8bdd1d">The Voice User Interface Design</a></li><li><a href="https://www.amazon.com/gp/product/1575860538/ref=as_li_tl?ie=UTF8&amp;camp=1789&amp;creative=9325&amp;creativeASIN=1575860538&amp;linkCode=as2&amp;tag=theuxfactor-20&amp;linkId=cbeabbb8ca8a88961be03c4517610dd7">The Media Equation</a></li><li><a href="https://developers.google.com/actions/design/conversation-repair">In Conversation, There Are No Errors</a></li></ul><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Ff%2F50d69a%3Fas_embed%3Dtrue&amp;dntp=1&amp;display_name=Upscribe&amp;url=https%3A%2F%2Fupscri.be%2F50d69a%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href">https://medium.com/media/05d5fd32eda31cbd1b83287606744532/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ef080c843640" width="1" height="1" alt=""><hr><p><a href="https://uxdesign.cc/voice-user-experience-design-and-prototyping-for-mere-mortals-ef080c843640">Designing Voice Experience</a> was originally published in <a href="https://uxdesign.cc/">UX Collective</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Lean Usability Testing]]></title>
            <link>https://medium.com/@arungeorge_/lean-usability-testing-5983baf4366b?source=rss-1737a51393c7------2</link>
            <guid isPermaLink="false">https://medium.com/p/5983baf4366b</guid>
            <category><![CDATA[arun-george]]></category>
            <category><![CDATA[usability-testing]]></category>
            <category><![CDATA[ux]]></category>
            <dc:creator><![CDATA[Arun George]]></dc:creator>
            <pubDate>Mon, 22 May 2017 13:52:27 GMT</pubDate>
            <atom:updated>2017-05-22T13:52:27.422Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Why Usability Testing?</strong></p><ul><li>Usability Testing can give deeper insights about your users and the product.</li><li>Testing users don’t have to be complicated or rigorous.</li><li>Testing users can be fun, easy, and most of all — FAST.</li></ul><p><strong>Conducting Lean Usability Testing</strong></p><ul><li>Recruit 5 users from your target audience</li><li>Select different tasks you want the users to achieve (change filter, log out, etc.) Test one user at a time.</li><li>Ask each user to talk-out-loud as he/she works on each task.</li></ul><p><strong>Observe user behaviors</strong></p><ul><li>To avoid bias, the researcher should follow a “script” each time when explaining tasks.</li><li>Analyze the results (number of clicks to navigate, different paths, facial expressions, etc.)</li></ul><p><strong>Why 5 users and not 500?</strong></p><p>“<em>Elaborate usability tests are a waste of resources. The best results come from testing no more than 5 users and running as many small tests as you can afford.</em>” — JAKOB NIELSEN</p><p>More @ <a href="http://ageorge.co/">http://ageorge.co</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5983baf4366b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>