Can AI react to players via voice input in a more natural manner? How can video games with AI characters increase immersion for players? Dive into these questions and more in this session from the live conference Pocket Gamer Connects Helsinki 2022 where Fortis Games game designer Eva Vital shares her views and more.
What is voice AI?
At the live conference Pocket Gamer Connects Helsinki 2022, Vital talks about the ins and outs of voice AI in video games. She shares its different aspects such as Automated Speech Recognition (ASR) – also known as Speech To Text (STT) – Natural Language Processing (NLP) and Voice Interaction Models. She also discusses how voice AI behaves to allow users to interact with software in a more natural manner.
There are various challenges that voice interaction designers need to account for when it comes to natural dialogue. Via intent-matching, designers can use sets of phrases that train the AI to understand certain intents. This includes “yeah” for “yes” and “nope” for “no”.
How intent-matching trains AI to respond to insults
Vital delves into how the new technology translates to games with multiple choices or those with text input. She discusses examples from the game “The 3% Challenge”, where players have the freedom to treat the AI characters badly or well. She also talks about how swearing is a natural part of dialogue. It makes the narrative less jarring and gives designers the opportunity to flesh out characters.
Toward the end of the session, Vital sheds light on the implications of using voice AI in games. There are benefits for movement and sight-impaired players; plus, the technology offers a more natural interaction and increased immersion for players. This 20-minute talk is ideal for anyone who’s eager to know about the ethics surrounding voice AI, such as zero tolerance for slurs and more. You can see conference panels like this in person by attending our upcoming partner events.