Google Working On Player Voice Interaction With NPCs In Games

"Google Play’s best apps and games of 2022" | Source: Google


  • Google was recently awarded a patent that aims to enhance the realism and immersion of video game dialogue by using a natural language model (NLM) to generate responses.
  • The system analyses language inputs such as dialogue selections, audio recordings, or text messages from players and identifies potential responses from a corpus of natural language candidate lines.
  • The selected response is presented to the player through the video game’s display or audio output, based on the highest probability of a match between the input and potential responses as calculated by the natural language model (NLM).
  • The patent suggests that the system can derive language input from audio within the video game, potentially enabling interactions in various languages and enhancing accessibility for players.
  • While the patent’s publication does not guarantee immediate implementation, if realised, the system has the potential to create more immersive and interactive video game experiences with a wider range of dialogue options for players.

Earlier today, we came across a recently published patent by Google titled, “NATURAL LANGUAGE DIALOG SYSTEM FOR VIDEO GAME INTERACTION,” filed in September 2020 under the name of GOOGLE LLC. The patent, published earlier this month, describes a system for video games that uses a natural language model (NLM) to provide dialogue responses. In simpler terms, it means that the system can understand and respond to the player’s dialogue or language input in a more natural and interactive way.

Block diagram of a game streaming system that employs a natural language model to select a dialogue response for a video game in accordance with some embodiments. | Source: Patent Public Search

“A video game system provides dialog responses based on a natural language model (NLM). The NLM is a language model that receives a language input, such as a dialog selection, audio recording, or natural language text input provided by a user of the video game system,” reads the abstract for the patent.

“In response to the language input, and based on a corpus of natural language candidate lines, the NLM identifies one or more potential responses. The video game system selects a final response from the identified potential responses and provides the selected response to the user via, for example, one or more display frames or via an audio output.”

The purpose of this system from Google is to make the interaction between the player and the video game’s characters or environment more realistic and immersive. It provides a wider range of options for dialogue input and allows the video game to respond in a more natural and dynamic way. This can enhance the overall gameplay experience and make it more enjoyable for the players.

When a player interacts with the video game, they can provide input through dialogue selections, audio recordings, or text messages. The system, which comprises a sophisticated natural language model (NLM), takes this input and analyses it. It looks at a collection of potential dialogue lines or responses stored in its memory.

Based on the player’s input, the system identifies one or more possible responses from its stored collection of dialogue lines. It tries to find the most suitable response that matches the player’s input. Once the best response is chosen, the video game system then presents it to the player through the video game’s display or audio output.

Flow diagram of a method of creating a dialogue response for a video game using a natural language model in accordance with some embodiments. | Source: Patent Public Search

According to the patent’s claims, determining the response involves using the natural language model (NLM) to calculate the probability of a match between the language input and a potential response. The response is determined based on the candidate with the highest match probability. Additionally, the match probability is determined by considering the language input and a specified constraint for the potential responses.

Moreover, the patent highlights an intriguing aspect by indicating that the language input can be derived from audio input within the video game, as well as from a player’s dialogue selection. Notably, the patent does not explicitly specify the language on which the natural language model (NLM) is trained. This intriguing omission implies that the system has the potential to facilitate interactions in various languages through audio input, providing a more accessible gaming experience for players who may not be fluent in English or prefer to communicate in their native language.

The implications of this patent are far-reaching. By enabling video games to interpret and respond to natural language inputs, developers can create more immersive and lifelike video game experiences. Players will have a broader range of dialogue options, allowing for deeper engagement and enhanced character interactions within virtual worlds.

Nevertheless, it is important to note that the publication of this patent does not guarantee its immediate implementation by Google or any other entity. The actual integration and utilisation of the proposed system within existing and future video game technologies will unfold with time. If realised, this groundbreaking system has the potential to deliver heightened realism and interactivity in dialogue exchanges, offering a wide array of possibilities for video game developers to craft captivating narratives and immersive gameplay experiences for players.

What do you think about this? Do tell us your opinions in the comments below!

Similar Reads: EA Allowing Players To Customise In-Game Character Animations

Was this helpful? 🕹️

Good job! Please give your positive feedback 😏

How could we improve this post? Please Help us. 💡