Spokestack Introduces Personalized Voice Assistant Development Platform for Web and Mobile Apps
Voice technology startup Spokestack has debuted a new tool for personalizing custom voice assistants using machine learning. The Spokestack Maker toolset aims at streamlining and speeding up the process of integrating custom voice assistants into web and mobile apps.
Spokestack Maker is designed to bring voice AI tools to developers and creators who are not necessarily experts in the technical aspects of the field. The goal is to eliminate those barriers to creating customized speech-based interfaces and allow those with ideas to play with the tech the way they would with a touchscreen or mouse-operated app. Users can set up wake words, vocabulary for the AI and generate the synthetic voice they want people to hear. The built-in machine learning algorithms gather the data as the voice AI operates and continually improves how well it performs. Using Spokestack Maker also skips the need to learn how to build software with Amazon Alexa, Google Assistant, or Microsoft Azure, which also helps save on development costs.
“Spokestack makes the state of the art in machine learning research accessible to developers with the full potential of voice technology, not the tiny pieces of it that the big vendors grant access to,” Spokestack president Noel Weichbrodt said. “There’s an unfilled hunger from the creator community for tools that make voice usable to them. The imaginations of individuals and startups are where innovation happens, and they are primed for Spokestack Maker to turn their ideas for voice into all kinds of exciting software applications.”
The toolset’s custom voices and keyword and wake word recognition have been operating in a closed beta until now but builds off of the existing custom voice features Spokestack offers for web interfaces and mobile apps. Spokestack released the initial customized voice for mobile apps in January. Then in June, the company added a complementary ability to make voice-enabled mobile apps perform like voice apps on smart speakers. Then in November, Spokestack introduced Tray, a more turnkey approach to making a voice assistant. Maker combines customization and speed for its users. Spokestack will host a guide toward using all of its customization tools on June 2 at 1 p.m. Eastern that you can register for here.
“Adding full-featured voice capabilities to applications can take months or even years of development effort,” Spokestack CTO Brent Spell said. “With Spokestack’s frameworks and services, voice features are as easy to build as the rest of the application. Spokestack Maker pushes this further by allowing developers to quickly prototype custom machine learning models for speech and language processing.”