[hemmerling] Smart Assistance - Digital Virtual Assistants, Digital Voice Assistants, Smart Assitants, Smart Speakers

AMAZON Alexa

The Service, the Tools

( Official ) Alexa Developer Resources

Third-Party Streaming Services for use with Alexa

  • Spotify "Alexa" - “Nutze die Sprachsteuerung, um deine Musik mit Amazon Alexa abzuspielen. Hinweis: Du benötigst ein Spotify Premium Konto, um Spotify auf Alexa zu nutzen:-(.

Design Concepts for Voice Applications

  • Idea ⇒ Contents ⇒> Design ⇒ Code

Third-Party Services for Application Development

Bottalk

  • BotTalk - “Alexa Skills and Google Actions with simple markup language”.

Invocable / Storyline

The Service ( 2017-09 - 2017-04-15 )
  • Invocable ( formerly: Storyline ) with free BASIC account, commercial PROFESSIONAL and ENTERPRISE accounts.
    • “Design and Prototype Alexa Skills. Create for the platforms of tomorrow. Today”.
    • “The easiest way to create an Amazon Alexa skill without coding”.
    • “Create Alexa skills without coding. There's no way to create Amazon Alexa skills unless you're a good engineer. Storyline makes it easy with a visual drag-n-drop interface”.
  • YouTube "Storyline" - “Create Alexa skills without coding”.
  • Medium "Hello, Invocable. Goodbye, clunky voice design tools", 2018-11-12 - “We’ve changed our name: Storyline is now Invocable”.
  • “The Invocable tool will be shutting down on July 1st, 2019. We've partnered with the team at Voiceflow to make this transition as easy as possible for our users”.
Business Analysis
    • At that time (September 2017) we had two assumptions in mind
      1. Voice apps are simple. If we build a platform that gives people an ability to create voice apps w/o coding, they’d love to use it.
      2. Non-tech creative folks are actually able to create good voice apps.
    • It took us a year to understand that we were wrong with our assumptions:
      1. Voice apps are not simple, they are hard. You need to be a great voice designer to build an engaging voice app.
      2. Non-tech creative folks are actually able to create good content, but the app is not just content. A skill is like a huge product on itself, rather than just a content piece. And most content creators are not good at product management.
    • For our product (DIY platform) to work, there has to be a simple, repeatable, and scalable use case like a landing page for websites. And we haven’t discovered one in the past 19 months.
      • DoItYourself ( DIY ).
    • That’s why we decided to pivot from a 'DIY platform' to 'Design tool for voice' in October 2018.
      • But after 2 months working on a design tool we understood that it doesn’t make sense: even though the smart speakers’ adoption snowballs, people are still using the same things on their Echo devices as four years ago — music, news, reminders, and other commands.
      • I said 'commands' because it’s exactly what a perfect voice interaction is — a very short, concise, and flexible command that can pass over a vast amount of information, including context and parameters. For example, 'order me the same pepperoni pizza I had a week ago'. We call them integrations, because the brain of this voice app is somewhere else — in a restaurant’s CRM, Uber’s backend, or NYT’s content management system. These voice apps have nothing to do with the design; it’s about getting the intents/utterances right. There’s nothing to 'design' here because the shorter it is, the better.
          • Natural Language Processing (NLP) refers to all systems that work together to handle end-to-end interactions between machines and humans in the preferred language of the human. In other words, NLP lets people and machines talk to each other 'naturally'.
          • Natural Language Understanding (NLU).
          • A good rule of thumb is to use the term NLU if you’re just talking about a machine’s ability to understand what we say.
    • To summarize:
      • The market of a tool for creating voice applications relies on the success of voice applications, which is not there yet.
      • A voice app works well as an integration — very short, concise request that is correctly recognized and processed. 'play the latest album from Eminem' is a good example. But there’s nothing to design here; all these applications are custom integrations, sometimes made by vertical players (like DoorDash skill for ordering).
      • There are voice apps that need to be designed, but NLP and NLU quality are not good enough now to support their growth. They’re like IVRs from the 90s, but on Alexa.

Voiceflow

  • Voiceflow - “Design, prototype, and build voice apps. Create voice apps for Amazon Alexa and Google Assistant. No coding needed”.

Anil Kumar & Alexey Vidanov

  • I participated in 2 sessions of Anil Kumar and Alexey Vidanov on the BarCamp UXCamp Hamburg on 2018-08-11 in Hamburg, Germany.
Anil Kumar
    • Step #1
        • Questionnaire.
          1. If you were a person out there in the world serving your customer, what is your job would be? Example: for Netflix, it would be Movies librarian
          2. Customer would describe you as the most? Examples for Amazon, it would Earth most customer centric company
          3. You never want to come off as? Examples for Netflix, it would be narrow
        • Your Product Personality
          • “If you were a person out there in the world serving your customer, your job would be 11111111. And customer would describe you as the most 22222222 of any in that profession. You never want to come off as 33333333”
    • Step #2.
      • What is your target audience? What does your target audience need which can be served by your Alexa-driven service?
    • Step #3
      • GitHub.io, Anil Kumar "VoiceX" for FireFox and Chrome browsers.
        • Choose Language - Many languages :-), but not German :-(.
        • Choose
          1. None
          2. Amazon Alexa
          3. Google Home
          4. Apple homepad
          5. Mobile
          6. Watch
          7. Male Avantar
          8. Female Avantar
        • Enter the text you want to get spoken by Alexa. Think of use cases, user stories, user experience.
        • Please switch to “Test mode on”
        • All entered text is now displayed and can be spoken by the browser application, by clicking on the HTML link.
        • Test the “voice application” with a user / customer. Let “VoiceX” speak Alexa's answers. Note missing answers, that should be added for a new test.
          • Customer: “Alexa, start “MyVoiceApplication”.
          • Alexa: ...
          • Customer: ...
          • Alexa: ...
Alexey Vidanov

Resources

  • Experts told me, that “AMAZON Echo Show” is a limited “AMAZON Fire” Android tablet.
  • Experts told me, that “AMAZON Fire” Android tablet need a(n AMAZON) proxy to connect to Internet.
  • Fortune "Amazon Is Converting Your Fire HD Tablet Into an Echo Show—for Free", 2018-06-28 - “The company on Thursday announced that it will begin rolling out a software update to Fire HD 8 and Fire HD 10 tablets next week called Show Mode. The feature delivers a hands-free Alexa service to the device that will display your calendar, allow you to watch video, check in on connected cameras, and more. The software update effectively converts Amazon’s tablets into Echo devices with screens. And whenever you want to go back and use them as a tablet, you can simply turn off Show Mode”.
  • The Verge
  • Trusted Reviews "Amazon Echo Show" - “Alexa is better with a screen”.
  • EN.Wikipedia "Amazon Echo Show" - “The Echo Show initially supported YouTube videos; on September 26, 2017, it was revealed that Google ... had blocked the device's access to the service, citing violations of its terms of service and ongoing negotiations. While Amazon later worked around the restriction by using the web version, Google announced that it would block YouTube from the Echo Show, as well as the Fire TV platform”.
    • “Ultimate Ears returns with another explosive duo of speakers, this time with Alexa”.
    • “It’s fun to ask Alexa about the weather, quiz it about a random fact, or perform one of the thousands of other Skills it has. Amazon's AI is also a decent DJ. The Blast and Megablast come with a 3-month trial of Amazon Music Unlimited, which was helpful since my music app of choice, Spotify, isn’t yet accessible via voice controls on these speakers. Other popular apps like Pandora are coming, but there's a wait for those as well.

APPLE Siri, APPLE HomePod

  • EN.Wikipedia "Siri", DE.Wikipedia "Siri (Software)" - “A virtual assistant part of Apple Inc.'s iOS, watchOS, macOS, HomePod, and tvOS operating systems”.
  • EN.Wikipedia "HomePod", DE.Wikipedia "HomePod" - “The device runs an iOS variation... It integrates Siri, which can be used to control the speaker and other HomeKit devices, and can be used to conduct text messaging and voice calls from an iPhone. The HomePod supports only proprietary Apple platforms and technologies”.

BAIDU "DuerOS" & BAIDU "Raven H"

GOOGLE Now, GOOGLE Assistant, GOOGLE Home

The Service

The Tools

Third-party Software

NPM "virtual-google-assistant"

Voiceflow

  • See on this page.

Third-partys Services

  • Bottalk - See on this page.

Resources

MICROSOFT Cortana

The Service, the Windows Tool, the Android App

Resources

SAMSUNG Bixby

Smart Home Devices which work with Digital Virtual ( Voice ) Assistants

Resources

Forums, Newsgroups

 
en/smartassist.html.txt · Last modified: 2019/10/17 19:49 (external edit) · []
Recent changes RSS feed Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki