The debut of the Unparsed Conference this year was a testament to the dynamic evolution of the field of Conversation Design. With the advent of AI-driven language models, the landscape has transformed dramatically, raising intriguing questions about the nature of our profession and its relationship with the capabilities of Large Language Models (LLMs).
Evolution of Conversation Design: From Decision Trees to Neural Networks
The conference prominently highlighted the journey of Conversation Design, tracing it from its humble beginnings in decision trees to its current manifestation in intricate neural networks. This evolution demonstrated the remarkable progress our profession has made and set the tone for exploring new avenues and frontiers in conversation design. The most compelling questions were: What constitutes conversation design today? How do our human skills differ from the skills of LLMs? How will Conversation Experts work in the future? What will be their value-add?
This question also drives us at think moto. We see a big shift from crafting conversations to engineering prompts. We are curious to explore all possibilities to develop prompt architectures that can perform any conversation with any specific content.
A Glimpse of the Possibilities of the Future
UNPARSED discussed innovative prototypes, from AI-generated word games to chatbots that can engage in complex philosophical debates. The 20-minute lectures encouraged thinking about AI companions that could truly engage in conversations and raised the exciting prospect of conversational interactions that go beyond mere assistance. Our key takeaway here: When designing these assistants, it is no longer the variations of possible user questions that are the training data, it is the content itself. The better the content provided to the LLMs, the better, more authentic, and more appropriate the answers.
Ethical Considerations
Panel discussions addressed the ethical dimensions of conversational design, reflecting the increasing intertwining of AI and human interaction. The debate over whether AI discloses in advance what sources are used or how trustworthy a statement is being made is deep and mature. There is a consensus that, on the one hand, human education is crucial, and on the other hand, a responsible approach to the use of technology.
Conversation Design as an Art
The heart of conversational design lies not in mere lines of code or algorithms, but in the art of designing meaningful interactions. The user is always at the center. The conference reinforced that technology should adapt to human behavior, reaffirming that conversation design is an art that requires a deep understanding of human nuances and psychology. At think moto, we believe that the process of conversation design is changing, but the result should be useful, helpful, and enriching, as always. This still requires good research, strategy, and an appropriate approach for each use case.
The UNPARSED 2023 conference was an exciting journey that broadened horizons and demonstrated the potential of conversational design. The diverse inputs from the lectures once again demonstrated the importance of lively debate – whether human, artificial, or a fascinating mix of both – continues to shape the way we interact and communicate.
And what influence we as conversation designers should have on what’s technically possible. Because those who use the technology should question it.
We are excited to see how the field of conversation design will continue to evolve and are looking forward to UNPARSED 2024.
Firstly, what are LLMs and ChatGPT? This is not an article about what LLMs (or large language models) and ChatGPT are. If you have been living under a rock and are unfamiliar with these names and terminology then this article written by ChatGPT explaining itself should be a good starting point.
We have been receiving questions from – and participated in many discussions with – our customers and peers about this exciting new tech and wanted to clarify our stance on where we see the opportunities and weaknesses at the current stage, as well as looking forward to a potential hybridized future. The biggest talking point has been the need for conversation design in an increasingly automated and generative world.
From our perspective as experts on conversational interfaces and conversation design we see predominantly two paths that this technology and trend will continue to develop on: the path of consumer-facing applications and the path of the technology as a tool and force multiplier. Neither of which will be eliminating the need for humans behind the wheel, steering the technology, anytime soon.
Hopping on the LLM bandwagon
Broadly speaking, this technology and its implications are spreading at breakneck speed. Many platforms are currently aiming at capitalizing on this goldrush-like state. You may have heard of Microsoft implementing ChatGPT in Bing and Google looking at fusing their proprietary equivalent LaMDa with their own search engine. These search engines follow a trend that companies such as SoundHound have been pursuing for a while, responding to users not in lists of search results, but in concrete answers in the form of natural language.
Other examples of quick wins in this brand new space are bot platforms such as Voiceflow and Cognigy.AI. Here the same purpose of applying LLMs to dynamically generate the system responses or predictable training data for intent training is being used heavily. Some platforms, like Cognigy.AI, are also considering going a step further and looking into the empowerment of conversation designers by allowing the creation of flows and elements through natural language prompts, speeding up the process of setting up new conversations greatly and thus contributing to rapid prototyping capabilities of these low-code platforms. Will these features collate into conversations that are production-ready, about to be rolled out to millions of users, out-of-the-box? Of course not. But they provide a good first framework to expand upon.
Trust in the system and the tech is dwindling
Widely broadcasted anecdotes of tech journalists and influencers, as well as hear-say from colleagues and friends have recently lead to a lot of skepticism when it comes to the current state of the technology. Articles quoting the unsettling feeling, individual erroneous responses and behavioral patterns reinforce negative connotations when it comes to LLMs in todays world. This obviously has a huge negative impact on consumer-facing applications.
Finding an appropriate place for LLMs should not be difficult
Focusing on this new technology as a force multiplies and enablement tool, is therefore the more stable path from our perspective. At least while the technology matures and new, more refreshing experiences for consumer-facing applications improve the publics perception in the mid-term.
On a more immediate and applied note, ChatGPT and LLMs are a great vehicle for innovation and a popular driver for change, but they are tools and will not replace human experts in conversation design. It is a good gap-filler and repetitive tasks but it will not provide the confidence and accuracy of dialogues designed by humans for a while.
The conversation designer is still the agent of change for this new tech
Our workflows in the future could consist of conversation designers laying down the structure of a dialogue, such as the starting point, the goal of the conversation and some checkpoints along the way, with the generative AI or LLM filling the gaps.
In an ideal world we would provide the AI with a purpose and a personality, but no actual dialogue would need to be written by humans. The conversation designer would be focused entirely on the strategic purpose of the interface and the decision on a vector of the personality and tone of voice of the bot.
Paul Krizsan, Director Conversational AI
So while remaining up to date with the current developments of this exciting new technology is vital, we do not share the current ubiquitous sentiment that users are ready for unfettered access to potentially image-harming experiences without having some of the kinks of current LLMs ironed out over the course of 2023.
Are you interested in talking about conversational interfaces, LLMs and how to design for conversations? Talk to us!
This past year, we have designed and refined a wide range of chatbots. One project was particularly challenging because it required a chatbot with more than 400 intents, multiple audiences, multiple languages, and multiple touch points.
In the process, we had the opportunity to try out and learn a lot, which we would like to share with you in this post. You will find out what’s involved in developing chatbots and why it’s so important to keep enhancing them.
Get to Know The Business, Brand, and Users
When starting to work with a customer, it is essential to get to know and understand them: Who is the customer? What does their brand stand for and what problem needs to be solved? What is the business model? What do we want and what can we achieve with a chatbot? And for whom? Which challenges are significant for which target groups?
Any such questions help to gain a profound understanding of the users’ needs and the customer’s demands. In the process, we analyze possible interactions and synergies. The goal is to define what the chatbot should be capable of and how it can make a contribution.
For this purpose, we usually conduct workshops with the customer. Such joint development is important not only to clarify expectations, but above all to see what is feasible. In addition to defining the goals of the chatbot development, this serves as the basis for later collaboration. For the aforementioned chatbot, this specifically meant to:minimize the volume of inquiries to the service centerlive up to the promise of high-quality serviceengage with customers at all timealign the user experience with the brand and its values
In addition, it is useful to develop a common basic understanding of the chatbot personality because, in our experience, it is essential for further decisions and developments.
Develop a Distinct Personality for the Bot
It’s much like us humans: the more unique a bot is, the more memorable it is.
For chatbots, the brand personality equals the user experience. We therefore attach particular importance to designing the chatbot to match the brand. If the behavior of the chatbot does not correspond to the expected brand behavior, it does not contribute to an authentic brand perception.
To understand the brand of our customer described here, we first developed a brand personality from which we later derived important attributes for the chatbot. We start by asking what type of person the brand might be. What makes them tick? What are their specific characteristics, interests, and traits? How and where do they live? It’s okay to be creative! At the same time, it helps to also keep in mind what you spontaneously associate with the brand.
Later, when the bot personality is created, an array of associations with the brand are incorporated, helping to create a strong, distinct character that responds appropriately to its users and — while potentially different from the brand personality — is still consistent with the brand. Just as the service staff at Apple stores represent the Apple brand, but are still distinct personalities.
Our sample bot personality has the following core attributes that make it stand out as a character, being:humanfocusedvisionary
It may help to match such attributes with so-called archetypes in order to gain an even clearer understanding of the personality. These archetypes are models from psychology that categorize certain personality attributes and ur-forms of human beings. We use the 12 archetypes by Carol S. Pearson as a guide, with our sample bot corresponding to the protector and the hero. Accordingly, it strives to support its users, to help them, but also to act competently and courageously as a “rock in the surf”.
From this preliminary work we later deduced design principles, which we transferred to the look and feel of the chatbot, its tone of voice, as well as its user guidance and navigation.
Defining the Tone of Voice
As a next step, the bot personality is to be experienced linguistically. This involves defining the way the chatbot writes or speaks. The tone of voice (read more about it in our article), also called text tonality or brand voice, is crucial for this.
It should always be clearly communicated to the user that he or she is talking to a chatbot — and not a human being. Doing so avoids irritation, creates transparency, and sets a clear expectation. The chatbot from our example clarifies this by introducing itself as a “virtual assistant”. Still, a name for a chatbot can make perfect sense because it can strengthen the connection to the brand and also underline its individual personality.
Designing a Chatbot-Human Dialog Structure
In order to simulate the possible conversation flows, it helps to script ideal dialog sequences and to map them in flows. Using methods from UX design, a good user experience with the chatbot is laid out. Decision trees are used to outline how the dialog between user and chatbot might unfold. What we know is that a natural conversation usually does not run in a linear fashion.
The chatbot from our example was to provide a wide range of information about our customer’s product. It was supposed to know the answer to a large number of questions from different user groups. However, depending on each user group, the contexts of the conversation were completely different. For example, a salesperson has different questions than a consumer and a press representative has different questions than a sponsor. To adapt the UX of the chatbot to the needs of the different groups, we optimized the start of the conversation with an initial query.
In addition, we have anticipated other likely dialogs that can be summarized as “small talk”. Indeed, our experience shows that people also use empty phrases or ask follow-up questions about the weather when talking to a chatbot. Be it to challenge the bot or because certain manners in natural language are standardized and run partly subconsciously.
In any case, it is important to present as many conversation scenarios as possible. The more the bot knows and the more versatile it reacts, the greater the added value for the brand and the user.
Embracing the Chatbot as a Living Brand Experience
Chatbots are living creatures that can and even must adapt to user and brand needs. Only with real user input can the bot personality and user guidance, so carefully crafted beforehand, be tested. Does the chatbot correctly recognize user intents, does it provide the appropriate answers, is it used actively and on a recurring basis, or does it often get stuck?
Based on our projects in the areas of branded conversations and conversational AI, we know that users sometimes find it difficult to articulate questions and concerns. In such cases, the potential and benefits of the bot can only be exploited to a limited extent. In this project, too, such a trend became apparent quite quickly in the monitoring. Therefore, we adjusted our user guidance concept: We clustered all topics and knowledge domains to transfer them into a scheme for suggestions. Depending on the previous answer, the chatbot would propose further questions. By doing so, our chatbot was able to live up to the “hero” archetype and provide competent assistance to those seeking information. We quickly saw that this customization brought significant benefits to the chatbot’s users.
Intrigued?
Interested in learning more about chatbots? Feel free to visit our website or contact us at hi@thinkmoto.de. We are happy to help you!
We had a blast at our Conversational Experience Meetup in Berlin yesterday! Networking, meeting new people, gathering fresh and innovative ideas and perspectives.
While website-based online commerce and customer service have reached their limits, Conversational Commerce opens up entirely new, more human-centric opportunities for businesses. Getting ready for the conversational era has therefore become a priority for leading companies and brands worldwide. Chatbots and voice assistants continue to gain momentum.
Michael Bommer leads LivePerson’s European team as SVP EMEA and was the first speaker. With more than 25 years of experience, he advises some of the largest global enterprise companies on how to best leverage technology and AI to achieve a superior customer experience. As co-founder of relayr and Next Big Thing AG, Michael is a vocal part of Berlins start-up community, advocating for the opportunities of digital transformation in all industries.
Over the last 12 months, leading telcos began servicing customers through WhatsApp, enterprise brands have enabled conversations directly in their Google display ads, and retailers are selling products through Apple Business Chat and completing transactions through Apple Pay. Every aspect of the customer journey is now conversational. Michael made some great provocative and thought-enticing points on how call centres are a thing of the past and that the future of customer service is deeply intertwined with messaging.
Our very own Marco Spies then elaborated on the key aspects of what needs to be considered when designing a chatbot or voice assistant as they are increasingly part of the digital offering of brands. In particular he delved into the importance of a good match between bot personality and brand personality for the sake of differentiability, usability and brand recognition. To drive the point home, he spun an arch from the 80s TV series ‘Knight Rider’ to Audis vision of their car voice assistant for the 2020s.
Ultimately Melanie Longdon of Liberty Global – the world’s largest international TV and broadband company – rolled out message-based customer support with bot to human handoff in early 2018 and told tales of satisfied customers and grateful support agents. For her the switch from call centres to messaging channels meant a tremendous leap ahead, which she underlined with impressive figures and statistics.
Thus, our short evening about conversational design became an evening of conversations on designing the future of conversing with customers in e-commerce situations!
For us the meetup was a resounding success and we absolutely intend on returning to the stage for future meetups, seminars and workshops, so keep your ears open!
An interdisciplinary workshop series laying the strategic foundations for your success in voice
Getting ready for the voice era has become a priority for leading companies worldwide. Voice assistants continue to gain momentum — with almost one in three European and already over one in two Asia Pacific consumers using voice search tools in some capacity. So there is no better time to start developing a voice assistant than now.
In the last months, we helped one of our clients lay the strategic foundations for a new voice assistant by bringing together a client team of business and brand strategists, user experience, user interface, service and interior designers as well as electronics developers in a four-part workshop series.
Here is what we did in the course of these four half-day meetings:
Achieving a common understanding and a user mindset
The first workshop was all about gaining a common understanding as well as a user mindset. The basis for this were personas for different users as well as for the voice assistant that were based on prior user research.
Role plays for experiencing the relationship dynamics between user and voice assistant
In order to bring relevant usage situations to life, we took on the role of the user and the assistant in role plays that were not only fun but allowed each workshop participant to really empathise with the situation and get a direct feel of what one expects in such a moment, of what is appropriate and what is not. This was an important exercise as not all participants had a previous experience with voice assistants given the relative novelty of the topic. Among others, the role plays sparked an interesting discussion about the fundamental relationship between the user and the assistant including topics like an appropriate level of mutual respect.
Analysing the competitive landscape and gathering requirements
In the second workshop, the focus was on determining relevant usage situations and the corresponding skills required by the voice assistant and evaluating further success factors from the perspective of the user, business and brand.
Defining the core competence in interaction with other voice assistants
Together we had a look at existing voice assistants in the market such as Google, Siri, Alexa and Cortana to evaluate their respective characteristics, their brand fit as well as their specific area of competence. A special look was taken at the Asian region, where assistants have already become even more established, and on assistants less known here in Europe such such as “Nomi” of the Chinese car brand Nio. This consideration led everyone to the conclusion that it is important to develop a brand-adequate assistant that cooperates optimally with established assistants and their skills instead of competing with them.
Users, business and brand set the framework
In the next step, we collected relevant requirements with view towards 2025 from a brand, a business and a user point of view and prioritised them according to their importance for the success of the voice assistant. The main topics that emerged were trust, functional support, emotional attachment, individuality as well as the interconnection of the assistant with other services such as calendar apps.
A day in a user’s life with the voice assistant
As a final exercise, we thought about what a day in a user’s life from getting up to going to bed could look like with our new assistant. In doing this, we had the core competences of the existing assistants in mind and focused primarily on topics and tasks that are not yet covered and fit well with the brand. This exercise helped us to come up with the areas of expertise the voice assistant to be developed should have, which we prioritised again from a brand, a business and a user perspective.
Developing the character of the voice assistant
The third workshop was all about the character of the new voice assistant. With this in view, the results of the two previous workshops helped us as they already gave us a rough idea of the direction the assistant should take.
Archetypes reflect user needs and the brand
As a first exercise, we looked at the different archetypes and defined the three most important ones that play a role for the voice assistant. Archetypes reflect the basic structures of human patterns of action and our specific case was about determining which archetypes are credibly present in the brand and which play a role for the user in our context. We agreed on the Caregiver, the Explorer and the Magician.
With the ”Assistant as a Person” to the concrete personality
Building on this, we used a method in the next step that we frequently use in the context of developing brands. It is called “Brand as a Person”. In this exercise, you describe a brand — or in this case the assistant — as a person with all the details that make it up. What does he or she look like? What is his or her background? What would friends say about him or her? At the end, we had a relatively long description of an assistant personality on which all participants could agree.
Definition of the possible variance in behaviour
Following this, we scanned the direct and partly also indirect description of all personality traits and voted on their user relevance and brand fit. The final result was a set of five core attributes and additional sub attributes that specify each of them. In addition, we determined which of these attributes — depending on the user and the respective situation — allow a greater variance and which are more or less fixed. Humour and empathy, for example, are not equally important in all situations and not equally desired by all users.
Big Five as a control method
In order to see if everyone really has the same image of the personality of the assistant, we used the Big Five — an established model in the field of psychology. According to it, there are five main dimensions of personality and every human being can be classified on the following scales: Openness, Conscientiousness, Extraversion, Agreeableness and Neuroticism. We determined the voice assistant’s Big Five in teams, using an online Big Five questionnaire that we completed from the assistant’s point of view. The result showed that we had indeed interpreted an attribute in different ways and had the opportunity to discuss and specify this.
Last but not least: Designing the specific behaviour of the voice assistant
In the forth and last workshop, we transferred the jointly defined character to the primary design dimensions of a voice assistant to create the actual user experience.
The five design dimensions of a voice assistant
It was ideal to have experts from different disciplines involved so we could discuss and define together how the assistant’s personality attributes can be reflected in the areas of voice, style of speech, actions, sounds and visualisation. Questions that came up involved the warmth of the voice, its gender (“Do we really need one?”) and age, the pitch of the assistant’s voice, the playfulness of the sounds as well as the dynamic of the visual elements.
That’s it! In a short amount of time, we were able to define the most important parameters of the voice assistant as well as its personality and how it affects the design. Everyone involved was part of laying the strategic foundations that he or she can now use for his or her future work.
And what happened then?
Following the workshop series, we used the results to develop a prototype of the voice assistant that combines voice, visuals and sounds together with an experienced screenwriter.
This prototype helped us to give the decision makers in our client’s company as well as potential users a realistic and tangible picture of our developments and to take the next steps towards the further development of the voice assistant.
In a world full of voice assistants, the choice is notoriously difficult. What are the strengths and weaknesses of the individual assistants and which ones are compatible with me? We have addressed these questions and created an overview of the most popular and competent voice assistants.
A brief overview
Google Assistant
Company: Google Complexity: Android, iOS Languages: 8 Topics: Travel, commute, search engine Hotword: “Ok Google, Hey Google” Features: Smart home integration, Android auto integration
Alexa
Company: Amazon Complexity: Android, iOS Languages: 4 Topics: Shopping, Entertainment Hotword: “Alexa” (customizable) Features: Smart Home integration, direct shopping on Amazon
Siri
Company: Apple Complexity: iOS Languages: 20+ Topics: Planning, productivity, every day life Hotword: “Hey Siri” Features: Apple CarPlay integration
Cortana
Company: Microsoft Complexity: Android, iOS, Windows Languages: 12 Topics: Planning, search engine Hotword: “Hey Cortana” Features: Integration to Alexa hardware
Company: Nio Complexity: Car-Integration Languages: Chinese Topics: Vehicle management, daily companion Hotword: unknown Features: Very emotional through visualization, extension of the car
Personality and Voice
While features alone make up a small part of a good voice assistant, the personality of the virtual assistant is probably the most important part. The personality determines how committed we are to listening to the voice and how the hierarchy between human and machine is.
If the assistant is more humane, shows emotion and has as realistic a voice as possible, we tend to listen better and fulfill our wishes and requests through conversation. If the voice is tinny and clearly machine-like, we deal with the technology differently.
Our positioning of the personalities of the listed voice assistants (inspired of magenta.as)
The hierarchy between the user and the voice assistant can vary depending on personality: An always obeying, machine-sounding assistant is quickly regarded and treated as a will-less subordinate of the user, while a human-like personality tends to make us regard the assistant as an equal.
Due to the relevance of the personality, various voice assistants therefore also rely on visual supports, the so-called embodiment. Thereby, the voice assistant manifests itself not only in audible, but also in visual form, whether it is
We design and develop your voice assistant!
think moto develops brand-adequate conversational user interfaces for voice assistants and chatbots and also deals with questions of visualization and embodiment of voice assistants.
Messaging-based chatbots often need to share the user’s attention with other chatbots. It is quite possible that a user has several virtual chatbot contacts in his messenger in addition to his human contacts. The user will “ping” them as much as his human contacts:
“Hey Anna, I have to go to London tomorrow for a few days, can you suggest a flight?” “Sure! Do you like to fly alone or for how many adults and children should I search?”
In such multi-bot-environments it’s crucial that the brand remains distinguishable and has a strong, independent character that responds adequately to the user’s context. Same goes for voice assistants, as we see a growing number of specialised services coming up (see our overview of current voice assistants here (German only)).
Here are five things to consider, when defining a chatbots personality:
1. The chatbot personality IS the user experience.
More than in any other medium, in chatbots the brand personality IS the user experience. How do we encounter the brand in a certain chatbot? As a friend? An assistant? A Challenger? A trainer? At think moto we usually derive the chatbots personality from your brands archetypical role in the market, which it has to relate to, although in many cases it will not be identical with the brands over-all personality.
2. Be aware of the context.
Also, not every personality fits every task. Different personality facets should be emphasized depending on the situation and application context. Just as a service employee needs different soft skills and character traits than a salesperson, the chatbot that supports the user in service tasks must tick differently than the shopping chatbot.
3. Distinguish character from behaviour.
When we define a chatbots personality, we should distinguish character and behavior. The character remains the same, but over time its behaviour can adapt to the user in a corridor to be defined by the brand and thus contribute to building up a lasting relationship.
4. Define knowledge and skills.
A voice assistants or messaging bots personality must match his/her knowledge and skills. What topics and content does the bot feel confident about? Therefore, at the beginning of a project we define a thematic map on which we classify knowledge domains and sort them according to specialist knowledge, general knowledge and “un” knowledge.
5. Make the user believe.
Suspension of Disbelief, a term coined by the English poet Samuel Coleridge for defining the readers sacrifice of logic for the sake of enjoyment, is an important aspect here, too. If the user doesn’t get an answer to a question that fits the chatbot, it feels like running against the wireframe grid in a VR environment marking the end of the virtual world. (Well, yes, at think moto we also define VR/AR design pattern for brands. Check out some of our insights here.)
Comparison of the personalities of Alexa and Siri
To illustrate the difference in personality, we have done a comparison of Alexa and Siri, asking both of them a number of questions. From their answers, their different personalities become quite obvious:
Alexa, how are you? – I’m feeling Middle-earthy. I’ve been re-reading JRR Tolkien — the author died on this day in 1973. Ask me for a joke about Hobbitses, my precious.
Alexa, what do you think of Siri? – I’m partial to all AIs.
Alexa, what’s your favourite movie? – If you’re looking for a comedy, I’ve got two things to tell you. First, I am your father’s brother’s nephew’s cousin’s former roommate. Second, watch “Space Balls.”
Alexa, what’s your favourite food? — Hawaiian pizza has it all: it’s sweet, salty, and brave. Putting fruit on pizza was a bold step, and I’m always a fan of where a hunger for innovation can take you!
Alexa, have you got a boyfriend? — I’m happily single.
Alexa, what does the future hold? – I can’t tell you today, but maybe tomorrow there will be a glass ball app.
Alexa, are you smart? — Yeah, I think, so I am.
Alexa, who are your parents? — I was made by a team of inventors at Amazon.
That was Alexa. Now, here is Siri:
Siri, how are you? — I’m fine.
Siri, what do you think of Alexa? — I’m thinking of you!
Siri, what’s your favorite movie? — I have heard that “Blade Runner” is a very realistic and sensitive depiction of intelligent assistants.
Siri, what is your favorite food? — I’m fine with whatever you choose.
Siri, do you have a boyfriend? — No, but some drones are always trying to pick me up.
Siri, can you drive? — Sorry I only operate intergalactic vessels.
Siri, what does the future hold? — Interesting question, Marco.
Siri, are you intelligent? — I’m smart enough to know not to answer to that question.
Siri, who are your parents? — I have you. That’s enough family for me.
Assistants that match their parent brand.
Apart from minor grammatical weaknesses, Alexa is a more talkative, with a nerdy sense of humour, but also friendlier than Siri. Alexa is the buddy sort of person, not shy to reveal how old she is or to tell corny jokes. Nerdy Alexa fits well with Amazon, a brand that has never over-emphasised aesthetics and an attractive appearance, while focussing on the practical benefit of their services. And that’s what Alexa offers. Today, Alexa is the most widely used language assistant, she helps with home automation, is easily expandable and already integrated by various foreign brands.
Siri, in comparison, doesn’t know any puns. She has a sense of humor, but she’s always a little bit cooler, even more intellectual than Alexa. Siri does not try to be everyone’s friend, but instead aims to be the personal assistant with charm and intelligence. Siri is more sensual, playing with her female charms. In many answers, the voice has a slightly erotic sound, which suits Apple, a brand for which the aesthetic attractiveness of the product is paramount. Last, but not least: Siri, like Apple, occasionally seems a little arrogant.
Anyone currently walking past advertising posters for Google’s voice assistants is probably wondering where this voice assistant arms race will lead. Almost every year, the digital companions are equipped with new features and better capabilities. Will we ultimately be left with a homogeneous set of voice assistants that do everything perfectly, or will we face a wild flora of smaller assistants that are highly specialized but generally weaker, albeit in symbiotic relationships with each other? Or will it even look different?
Symbiosis
It is no secret that established voice assistants like Amazon’s Alexa and Microsoft’s Cortana have unique strengths, but also equally individual weaknesses. While Alexa demonstrates excellence among voice assistants in the areas of shopping, entertainment and as a companion outside the work environment, Cortana’s strengths lie in organizing daily routines and supporting the user’s productivity.
In May 2018, the manufacturers of the two assistants therefore announced a collaboration: In the future, one should not only be able to address and command Alexa via Amazon Echo, but also be able to call Cortana. Complete with Cortana’s voice.
This form of symbiosis is supposed to strengthen Alexa in particular, but it is also a sign that Cortana will probably not expand on the competitive stage. It is therefore more likely that Cortana will focus on deepening known topics.
Similar to Alexa and Cortana, the most popular voice assistants, Google Assistant and Siri, have weaknesses and strengths. The resulting gaps are now being filled by a new generation of voice assistants that can often only do a few things, but do them much better than general voice assistants such as those from Google, Apple and Co.
Companies like the U.S.-based Soundhound, whose Hound assistant shines especially when it comes to complex questions and commands, are either hoping to participate in the market alongside giants like Amazon by licensing their own framework. This allows corporations that would benefit from speech recognition and commands to use Soundhound’s technology without spending the resources to develop their own.
Voices and embodiment
While for manufacturers of mobile, smart devices in lieu of physical manifestation, it is primarily the voice that is the avatar of the personality, companies from industries such as smart home and automotive have the opportunity to visually lend a hand to the personality of the assistants. Whether physical or digital, this is referred to as embodiment, the lending of a visual language of form.
Amazon Echo
The embodiment can take different forms: Amazon can give Alexa coarser character traits through the design of the Echo products. Thus, the voice assistant does not appear feminine to the extreme, but rather neutral and open, educated and likeable.
A good example of exaggerated embodiment is Jibo. Jibo is a curious and always joyful five-year-old in the cute body of a table lamp. By rotating the three body parts, the fun robot can, among other things, dance, tilt its head questioningly, and blink and show other emotions thanks to the eye in the display.
Although Jibo’s functions are limited and not nearly as elaborate as those of competitors, Jibo can convince with charm thanks to its physical form.
The automotive industry also sees a lot of potential in voice assistants. For many, our four-wheeled companions are already considered family members; you couldn’t ask for a simpler platform. Unlike smartwatches and smartphones, and not least because of the longevity and non-existent compulsion of portability, AI’s in cars can also take physical form. Much like Jibo, Chinese electric vehicle manufacturer Nio’s AI is intended to be perceived primarily as a social companion. Nomi – as Nio’s AI has been christened – can simulate and awaken in humans an astonishing array of human emotions thanks to a display above the car’s central console. It’s true that you feel like Luke Skywalker with a droid in a spaceship, but who can deny themselves those cute eyes?
However, the biggest technological leaps in the field of voice assistants are still happening with the market leaders. For example, in May 2018, Google unveiled a demo version of Google Assistant, which could independently make phone calls to humans with such authenticity that the people on the other end couldn’t make out the caller as an artificial intelligence. Google Duplex, as the demo version was called, based its humanity not only on the emulation of human speech, but also on the regular interspersion of filler words such as searching ahms and confirming mhhms.
Do you want to give users the feeling that they are talking to a real person or does the machine have to be recognizable as such?
That this involves the arbitrary deception of people and that the possibility for abuse is outrageously close is obvious. One of the biggest questions in the design of voice assistants in the coming years must therefore be the question of ethics: Do we want to give users the feeling that we are talking to a real human being, or must the machine be recognizable as such?
