Researchers at Carnegie Mellon University are developing a video game called “The Rolling Rhapsody”. The team developed the game with the mission of garnering in-home sounds of players while they play. They are trying to build an open-source database of the same so that they can use those sounds to train artificial intelligence in the near future.
Why is the Purpose of Rolling Rhapsody?
A voice integrated smart device can always benefit from these noises, as it would know what to ignore. More often than not, we do find reviews and reports from customers complaining about how their smoke alarms go off without a cause, or the smart security cameras send alerts because a dog ran down the stairs.
The goal here is to collect all the normal, ordinary noises that occur in a household every day. Followed by feeding it into the AI accordingly to make it differentiate better, and hence function better too.
A recent experiment conducted also proved that smart speakers are always up – without a wake word activation. Companies, however, are trying to release upgrades and updates to curb this.
For example, Amazon is upping Alexa’s security requirements, and Google allows its users to assess wake word sensitivity.
What are the Rules to Play Rolling Rhapsody?
Rolling Rhapsody is anything but complicated. All the players need to do is roll a ball around the screen, to collect treasures from the pirate base. Meanwhile, before signing up, they have to agree to submit recordings of sounds; be it on their iOS device or their android device.
The game grants the user the flexibility to delete any recording that might have been personal. While playing, the player and anyone watching the live stream can hear all the recordings.
The developers mention each and every sound is important – from the creak of a chair to the sound of a sigh, every sound can make a big difference. For now, the players need to manually label the sounds, which makes it prone to humane error.
However, it’s letting the developers collect a ton of sounds at the same time, so we think that can pass.
CMU Professor Jessica Hammer’s Statement on Rolling Rhapsody
CMU Professor Jessica Hammer said in a statement, “This data could be used to create extremely useful technologies. For example, if AI can detect a loud thud coming from my daughter’s room, it could wake me up.
It can notify me if my dryer sounds different and I need to change the lint trap, or it can create an alert if it hears someone who can’t stop coughing.
We can use this as a proof of concept for a new kind of game experience that can result in ethical data collection from the home. We can collect data in a way that’s fun and feels good for everybody involved.”
Does Rolling Rhapsody Pose a Threat to Privacy?
This livestream sound project is a joint venture, in collaboration with Polyphonic. CMY also homes researchers in Apple who are working on the Listen Learner project, to train AI with minimal training.
However, the Listen Learner project works way differently than CMU’s. It consists of microphones placed all over, and always on. It questions the users what kind of different sounds it can hear so that it can label them for future reference.
Even though this whole strategy might be fruitful, it surely does abuse the privacy of a user, as the mic is always on.
And this isn’t the first time that smart speakers have been accused of recording conversations without any authorization. It is what has resulted into the companies releasing upgrades time and again, to resolve the whole privacy fiasco.
Voice War
But voice technology is ever-growing, and the developments in this industry are skyrocketing. It’s just the beginning with a few stumbles, we still feel voice technology shall emerge victorious in the long run nonetheless. Mozilla is also building what’s said to be the largest database. Only time will declare the one who wins!
