ad
ad

Why Obsidian uses AI voices for game development | Sonantic

People & Blogs

Why Obsidian uses AI voices for game development | Sonantic

Introduction

In the rapidly evolving world of game development, dialogue plays a crucial role in making characters feel believable and ensuring that the player's experience is immersive. For Obsidian Entertainment, a developer known for their story-rich games, achieving such realism in dialogue has always been a priority. To achieve this, they have integrated advanced text-to-speech (TTS) technology from Sonantic into their workflow. This article delves into how this integration has transformed Obsidian's development process and highlights the potential it holds for the future of game design.

The Importance of Dialogue

Justin Bell, the Audio Director at Obsidian, emphasizes the significance of dialogue in their games. "The type of game that Obsidian makes has a lot of dialogue," Bell explains. "When you interact with the many characters that we have in our games, we want them to feel like believable people." The pacing of the game, character reactions, and emotional impact are all critical factors that are meticulously crafted to meet player expectations.

Traditional vs. Advanced TTS Solutions

Previously, Obsidian relied on basic TTS solutions that offered a monotonous and robotic performance. This was far from the final product that would feature actual actors, but it provided a rough idea of how the dialogue would fit within the game's context. The limitations of such basic TTS were apparent, and the team had to constantly remind themselves that the final voiceover would sound significantly better.

However, Sonantic's advanced TTS solution marked a massive improvement. "The difference between the previous text-to-speech solution was really night and day for us," Bell notes. Sonantic's technology allows Obsidian to evaluate not just the sound but also the appropriateness of emotional expression, pacing, and tone. This ability to iterate with a realistic approximation of the final product has been a game-changer.

Advanced Features

Sonantic continually evolves its technology to offer new features that further enhance the development process. During a call with the AI team, Bell experienced a demonstration of the new "expressive shouting" feature. He could adjust the style and intensity of the voice, like making a character shout convincingly. Such features provide nuances and emotional heights that closely resemble a Hollywood actor's performance.

Realistic Performances

Before entering the tech world, Sonantic's team members had experience in the film industry, working on major films such as "Harry Potter" and "The Dark Knight." This cinematic background drives their ambition to capture nuanced and varied performances in TTS technology. The aim is to bring the quality one would expect from a Hollywood actor into the realm of gaming.

Impact on Studios and Actors

While TTS technology is valuable for iteration during game development, it doesn't replace the need for professional voice actors. "We always go to the studio and record voiceover with actors," Bell confirms. Sonantic's TTS comes into play during the iterative stages, allowing developers to hear how the dialogue will sound when delivered by an actor.

Sonantic ensures that professional voice actors benefit from this technology. They offer a profit-sharing model for actors whose voices are used, preserving their voice quality and allowing them to work on multiple projects simultaneously. This arrangement creates a consistent passive income for actors without straining their voices.

Efficiency in Dialogue Management

Obsidian's games feature tens of thousands of lines of dialogue, making manual adjustments impractical. Sonantic's API allows developers to process these massive volumes efficiently. "Synantic allows us to just take those tens of thousands of lines of dialogue and just send it to the API," Bell explains, ensuring that what they get back is something that sounds real. This capacity to handle large-scale dialogue with realistic outputs represents a significant advancement in game development.

Conclusion

Obsidian Entertainment's adoption of Sonantic's advanced TTS technology showcases the potential of AI in transforming game development. The ability to iterate with near-final quality dialogue not only streamlines the development process but also ensures that the final product meets high standards of emotional resonance and realism. As TTS technology continues to evolve, it promises to further revolutionize the way stories are told in video games.

Keyword

FAQ

Q: Why does dialogue play a crucial role in Obsidian's games? A: Dialogue is essential in making characters feel believable and ensuring that the player's experience is immersive.

Q: What were the limitations of the basic TTS solutions used previously by Obsidian? A: Basic TTS solutions provided a monotonous and robotic performance, far from the final product featuring actual actors.

Q: How has Sonantic improved the TTS experience for Obsidian? A: Sonantic's advanced TTS solution allows for evaluating emotional expression, pacing, and tone, providing a more realistic approximation of the final product.

Q: What unique features does Sonantic offer? A: Sonantic offers features like "expressive shouting," allowing for nuanced and varied performances resembling those of Hollywood actors.

Q: How does Sonantic ensure professional voice actors benefit from TTS technology? A: Sonantic offers a profit-sharing model, preserving voice quality and allowing actors to work on multiple projects while earning consistent passive income.

Q: How does Sonantic's API facilitate dialogue management for Obsidian's games? A: The API allows Obsidian to process tens of thousands of lines efficiently, providing outputs that sound realistic and ensuring high standards of emotional resonance and realism.