Powder, an AI-powered cutting tool for games, can detect when a creator is shouting during a stream

Image credits: powder

powder, an AI-powered cutting software that takes highlights from game streams and turns them into short videos, will soon be able to detect players shouting to create better montages. The platform is also working on speech-to-text software so creators can grab a transcript of the entire broadcast and search for keywords.

Powder has developed more than 40 gaming-specific AI models, including audio analysis and laughter detection, as well as… Independent models For popular titles like Fortnite, Valorant, Apex Legends, Call of Duty, Rocket League, Fall Guys, Elden Ring, and Among Us. The company is also launching a Counter-Strike 2 mod.

All models work similarly. The AI ​​scans streaming recordings — whether from Twitch, YouTube, or an MP4 file — and finds spikes in activity, including wins, assists, kills, and other in-game performance-based moments. Powder takes these highlights and creates short montages for creators to upload to social media.

Image credits: powder

Similar to its ability to recognize laughter, the platform will soon launch another AI tool that recognizes fluctuations in audio so creators can create clips of them screaming — a common reaction when playing intense ranked matches. The company expects to launch it in mid-December.

“From uncontrollable laughter and rage to stopping even when there’s nothing obvious happening on screen, the best moments in games are so subjective and need to be reflected in several different perspectives that extend beyond the gameplay itself, “Barthelemy, co-founder and CEO of Powder Case, told TechCrunch. “This made us sure that we needed to capture the emotions of playing games with your community. This combination of skill-based moments and deeply emotional moments is what makes creating gaming content so unique and special.

Speech-to-text technology will also be available on the platform next month, giving creators a transcript of the broadcast and enabling them to quickly search for specific words and show the best highlights. Streamers can also enter mood prompts. For example, “Find five funny clips that make my fans crazy.” The software is designed using the player’s language to help make the results more precise and accurate.

“The ability to search for clips and place them in the context of long-form videos like Twitch streams using AI is the holy grail for content creators and the teams that support them, from video editors to their agents and managers,” Case said.

Image credits: powder

Additionally, Powder is updating its “Community Hype” feature, which will be rolled out next week. The AI ​​model was launched in September and detects chat spikes. The update will recommend clips where “the community is going crazy,” Case said.

“The release of the second phase of Community Noise Detection aims to open another perspective that influences what makes a ‘moment’ in a stream. One dimension of this is understanding what the community, who knows who is streaming best, is thinking. Communities have a keen sense of what matters in a gaming session Or a specific broadcast. “In this latest release, when the community goes crazy and wants to remember a moment, that’s the moment Powder AI will recommend you keep as a clip to share,” he explained.

According to a Powder survey of more than 3,200 streamers, creators spend an average of 53 hours per month or 630 hours per year researching and editing highlights. The powder claims to save streamers over 10 hours per week or 520 hours per year.

The France-based startup was founded in 2018 by Kees, Yannis Mangmatin, and Christian Navello. It has raised $22 million so far.