Jovo v3 Launches with Support for More Platforms, More Devices, and Custom App Experiences
Jovo v3 launched this week and is a substantial upgrade for the leading open source voice app development framework. Past versions enabled developers to create a single voice app code base and publish it simultaneously to Amazon Alexa and Google Assistant. Jovo v3 extends that publishing capability from a single code base across more platforms (Samsung Bixby, Facebook Messenger, and Twilio Autopilot), devices (Raspberry Pi and other hardware), and app experiences (mobile and web apps).
Jan König, co-founder and CEO of Jovo, says “Jovo has become the most popular open source framework for cross-platform voice apps.” He added that developers have created over 20,000 projects with the software and that Jovo-powered voice apps are generating over 10 million interactions per month. The new features have the potential to expand the market of Jovo use significantly.
Jovo’s key focus is a write-once and run anywhere voice development framework that enables voice experiences can be continuous, consistent, and complementary across devices and platforms. Developers don’t want to offer users a different experience on Alexa, in an iOS app and on their website. They not only want consistency in code base to reduce development and maintenance efforts but also want to deliver a recognizably consistent user experience. Each user environment has different capabilities and limitations so Jovo cannot make them completely identical but a common framework can minimize the variation.
Going Beyond Consumer Voice Platform Support
The expansion to support Bixby, Messenger, and Twilio will certainly be welcomed by developers that want to maximize their user reach without having to maintain numerous code base variants. However, the more significant expansion is to custom voice assistant environments. This is important because adding voice to web and mobile apps is an important trend among large consumer brands, media organizations, and enterprise companies. König commented:
In v2, the focus was still on Alexa and Google Assistant. We now offer the ability to integrate with ASR, NLU , and STT tools so that companies can add voice interactions to any platform or device.
König borrows a term I use often in my presentations about the state of voice assistants calling custom assistants, “the most important trend in voice.” He wrote in a post published this week, “When we started working on Jovo, voice was mostly seen as a platform. Companies built apps on top of the major voice assistants mostly for marketing reasons: they wanted to reach new potential customers through Alexa and Google Assistant…When voice becomes one of the most important elements of human-machine interactions, do companies really want to put all their fate in the hands of a few gatekeepers? We’re already seeing that companies are starting to build their own specialized assistants.”
A Ticket to RIDR
A key feature in Jovo v3 that enables custom voice assistant development is the RIDR Pipeline. It breaks voice app functions into three high-level categories: interpretation, dialog management, and response. These then support modular features such as automated speech recognition (ASR), natural language understanding (NLU), integration with content management systems, and text-to-speech (TTS) natural language generation. Regardless of whether you are building for Alexa, Google Assistant, or your iOS app, you can treat all of these as services and call the appropriate one based on platform, device, and context.
This doesn’t remove all of the complexity of building for multiple platforms. However, it is different than building each from the ground up in different SDKs.
Developer Reception is Positive
The initial developer response has been very positive. When developers realized they would likely need to support at least two platforms for their voice apps (Alexa and Google Assistant), they begrudgingly acquiesced. However, as more platforms have emerged and interest in publishing to mobile apps has increased the complexity has risen significantly. Jovo is one of the few options out there for developers to maintain control over their code base and enable support for multiple platforms simultaneously.
Still wrapping my head around how epic this update is. I switched all my development to Jovo last summer and can’t imagine working without it now.@jovotech is the gift that keeps on giving – thanks Jan & team for seeing the forest for the trees and pushing #VoiceFirst forward https://t.co/0tMxAmS2vr
— John Gillilan (@bondad) February 28, 2020
Congrats! The hardware part is truly impressive!
— Gian Segato (@giansegato) February 28, 2020
Thank you for all the work you do to make developing for voice so much easier.
— Tom Hewitson (@tomhewitson) February 27, 2020
John Gillilan, founder of Bondad.fm and an Alexa champion offered additional insight behind developer enthusiasm in saying, “Voice is so much bigger than one platform. Over time, voice as a modality will be woven throughout all types of products, experiences, and environments – from the voice platforms of today to mobile apps and websites, to cars, TVs, and hearables – and even custom hardware integrated into public spaces over time. As a voice creator, centralizing development will become increasingly important to nimbly explore all these new places. Developing for multiple voice platforms is messy.”