Please enter your username or the email address associated with the account so we can help you reset your password.
Think about your last great conversation. What made it special? Maybe it was the other person. Or maybe it was the fact that the conversation felt natural and was something you could get lost in? Were you riding the highs and lows of whatever story you were being told, forgetting about everything and everyone else around you? Probably so, and that’s the great part about great conversation. It’s immersive.
The types of conversations you create for a voice application shouldn't be any different.
Amazing conversational experiences are not only immersive, they’re also expressive. Through conversational AI, we can now move far beyond basic voice commandsand make the shift toward meaningful conversations with voice assistant powered devices.
Expressive conversations bring about unique character, voice, and highly designed sound environments. They do not just rely on the default Text-to-Speech voice of Alexa or Google Assistant. Voice conversations are meant to create unique, differentiated and memorable experiences, and this is critical to building brand loyalty.
Neuroscientists have found that voice and sound are the most effective drivers of emotion because they work underneath our cognitive radar. Expressive voice application experiences represent a unique opportunity for brands to create a strong emotional bond with customers. By exhibiting personality through conversational AI, brands connect with customers in a new way.
Voice Apps and Unexpected Voice Commands
To design an impactful conversation, the voice application must prepare for the unexpected. It’s part of the Voice User Interface (VUI) design process, and that process begins once a voice application is ready to answer all the expected use commands or questions.
If answering the expected is all an app can do, it isn’t creating an experience. It’s just doing what any app on any device could do, except it’s controlled by voice. And, while that’s fine if you want to check the weather or get a balance from your bank account, it’s not the kind of experience that will create an emotional bond with an app or a brand.
VUI design is all about handling the unexpected. How will your app respond and in what tone? The ensemble of all replies to unexpected requests define the true personality of your application. Just like in real life, people will remember and re-engage more frequently with apps that have a strong personality.
However, until now, creating expressive apps with custom voices, advanced sound design or the ability to handle a large amount of unexpected requests was very complex both to create and maintain over time.
Converse 2.0 changes that. Converse 2.0 empowers VUI designers and developers to imagine and craft immersive and expressive voice apps that showcase your brand personality through conversational AI.
Creation of Immersive Sound Experiences
Creating voice apps that include immersive sound design and recording of voice talents add to the complexity of voice app development. Lines need to go through final approval before sending to a recording studio (recording voice talent is not cheap!). Thousands of audio files need to be tracked and thousands of existing Text to Speech lines need to be replaced with audio files. Converse 2.0 eliminates these two challenges and makes it easier than ever before to create expressive and immersive voice applications that provide unique and memorable experiences driving usage and re-engagement.
Converse 2.0 introduces a new view into your voice project that lists all the response lines and groups them into logical filters. The lines can be selected and exported in order to be approved for brand or legal compliance. Does your legal department require that one line be updated? No problem, from our new Line Explorer view, you can directly edit the line without having to navigate through the logic flow of the project.
Dialog lines can also be exported to be sent to a recording studio. There, voice talents record each of the lines as a separate audio file. That can amount to hundreds or thousands of files. Converse enables sound designers to batch import all the audio files and automatically associates them with respective dialog lines in the project. If a specific line needs to be re-recorded, Converse makes it super easy to remove or swap the audio file for that line.
Building Dynamic Conversational AI
In French, when someone abruptly changes the subject of a conversation, we say “passer du coq a l’ane”, literally, to pass from the rooster to the donkey. In Converse, we call these type of non-sequitur changes, interjections. The ability to handle changes in context, or interjections, differentiates voice applications from IVR trees.
Let’s take a banking example.
App: You are about to pay your monthly cable bill for $120. Do you want to proceed?
User: Wait. What is my account balance?
App: The account balance of your checking account is $623.
App: You were about to pay your monthly cable bill for $120. Do you want to proceed?
“What is my account balance” constitutes an interjection. It was not one of the expected answers by the previous prompt.
“You were about to pay your monthly cable bill …” constitutes a segue and brings the user back to the flow of the conversation.
Converse 2.0 makes it very easy for VUI designers and developers to define multiple interjection sets that will listen for user interjections at different places in the conversation. For example, the answer to a request for help may vary depending of where the user is in the voice application. For each interjection, Converse also enables designers to define how to segue back into the conversational flow.
Combined with our existing ability to handle fallbacks (when the voice app does not understand what the user said), interjections and segues provide VUI designers with limitless creativity to make computer conversations sound more natural and shape a unique personality into their voice app.
Simplifying the Voice App Design Process
Converse provides a visual interface that makes it easy for non-developers to create the flow designs of a voice application. Visual interfaces are a great to get started, but as the project complexity increasestoward conversational AI, the representation of hundreds of nodes become a challenge.
In Converse 2.0, we are solving this challenge with the introduction of Modules. At their core, Modules enable you to collapse multiple conversation blocks together to save space on the top-level Canvas.
In addition, Modules have some very powerful properties: they are reusable, they serve as a scope for interjection sets, and they can be nested.
Finally, modules can be used as initial placeholders when drafting the VUI, which can then be built-up as the project evolves. With modules, Converse 2.0 allows the creation of large and complex voice apps in a graphical and easy to use interface.
Great voice apps are the ones that users come back to again and again. Building that kind of loyalty and brand for your app requires more than great functionality, it requires an immersive experience on a conversational AI platform. Those memorable experiences require an app that expresses your brand’s unique personality, character, and voice. Converse 2.0 provides unmatched flexibility for cross-functional teams to imagine and create those immersive experiences, like the award- winning WestWorld skill.
View Original Article