ReadSpeaker Debuts Voice User Interface Platform for Nintendo Switch
Digital voice interface creator ReadSpeaker has launched a tool for adding a voice user interface (VUI) to the Nintendo Switch video game system. Game developers can apply ReadSpeaker’s software development kit to embed the company’s deep neural network text-to-speech engine into their products, opening up a new dimension for the Switch’s library of games.
The Nintendo Switch is a remarkably flexible system, named for its combination of mobile and home-based playing. The game developers are sometimes more limited compared to other PC or other console platforms. ReadSpeaker simplifies adding a VUI to a game. The developers can add real-time text-to-speech to their game, pulling voices from ReadSpeaker’s stable for conversations and announcements in a game. They can even generate a custom voice from an actor to integrate into their product. With so much text in modern games, adding voices makes them a lot more accessible to younger players.
ReadSpeaker’s SDK also runs entirely on the device. That’s a significant benefit compared to the standard model relying on a cloud server. The voice elements of games won’t need to be connected to the internet to work, which gives them the ability to function when the Switch is played as a mobile device and not plugged in at a player’s home. The SDK is set up for global release, offering 11 languages and dialects for translation. ReadSpeaker’s deep neural network works to make the synthetic voice sound more lifelike and emotional. The developers can make their voices sound happy, sad, upset, or whatever else is appropriate for the part of the game.
“As recently as a few years ago, the quality of TTS solutions did not meet the exacting standards of highly produced, big-budget games. Digital voices at the time could not accurately represent different character types (and differentiate an evil character from a silly character, etc.) and as a result TTS has long suffered from an image problem – until recently there was a stigma that TTS voices were too synthetic,” ReadSpeaker marketing director Nate Murray explained in a blog post about how the tech works with video games. “With today’s neural voices however, characters can be voiced on-brand with consistent quality and speaking style. Emotion, laughter, and other para-linguistic sounds and expressions combine to bring synthetic characters to life and stretch the realm of possibility beyond what was available just a short time ago.”
ReadSpeaker’s new Nintendo offering is the latest in an eclectic mix of new partnerships and services from the startup. Spotify recently picked ReadSpeaker as the interactive voice for its new Car Thing automotive audio player. That arrangement became public soon after the U.S. military picked the startup to help train pilots to fly the F-35 Lightning II fighter jet. The company’s synthetic voice engine has attracted interest from many enterprises, including SoundHound, recently set on the path toward going public. SoundHound teamed up with ReadSpeaker last year to give its Houndify voice AI platform more lifelike, distinct voices.