Powder, an AI clipping tool for games, can detect when a creator shouts during a stream

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

Powder has developed over 40 proprietary game-specific AI models, including audio analysis and laughter detection, as well as standalone 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 model for Counter-Strike 2.

All models work the same way; AI analyzes stream recordings – whether from Twitch, YouTube or an MP4 file – and detects spikes in activity, including wins, assists, kills and other gameplay-based moments on performance. Powder takes these highlights and creates short montages that creators can upload to social media.

Image credits: Powder

Similar to its laugh recognition ability, the platform will soon launch another AI tool that recognizes voice fluctuations so creators can generate clips of them shouting – a common reaction during intense ranked matches. The company plans a mid-December launch.

“Uncontrollable laughter and rage stops until nothing obvious is happening on screen, the best moments where the game is very subjective and must be reflected with several different perspectives that extend beyond of the gameplay itself”, Barthélémy, co-founder and CEO of Powder. Kiss told TechCrunch. “This convinced us that we needed to capture the emotion of playing games with your community. This combination of skill-based moments and deeply emotional moments is what makes gaming content creation so unique and special.

Text-to-speech technology will also arrive on the platform next month, giving creators a transcription of a stream and allowing them to quickly search for specific words and extract the best highlights. Streamers can also enter mood prompts. For example, “Find me five funny clips where my fans go crazy.” The software is tailored to player lingo to help make results more precise and precise.

“Being able to search and contextualize clips in long videos like Twitch streams with AI is the holy grail for content creators and the teams that support them, from their video editors to their agents and managers,” he said. Kiss said.

Additionally, Powder is updating its “Community Hype” feature, which will roll 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,” Kiss said.

“The release of the second phase of Community Hype detection aims to open up another perspective on what constitutes a “highlight” in a stream. One dimension of this is understanding what the community thinks, who knows a streamer best. Communities have a great sense of what matters in a given game session or flow. In this latest version, when the community goes crazy and wants to remember a moment, that’s a moment that Powder AI will recommend you keep as a clip to share,” he explained.

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

The France-based startup was founded in 2018 by Kiss, Yannis Mangematin and Christian Navelot. To date, $22 million has been raised.