The Amazon-owned platform for sharing video gameplay and commentary said it has updated filtering tools to better detect hate speech in chat forums, with more improvements on the way.

Hate raids involve overwhelming streamer comment boards with abusive remarks, sometimes using automated or “bot” accounts.

“We’ve seen a lot of conversation about botting, hate raids, and other forms of harassment targeting marginalised creators,” Twitch said in a string of tweets.

“You’re asking us to do better, and we know we need to do more to address these issues.”

The posts came in response to complaints about racist, sexist, and anti-gay onslaughts aimed at Twitch streamers who don’t appear to be straight, white and male.

People targeted with such abuses shared their viewpoints and stories on Twitter with the hashtag #TwitchDoBetter. Among them was Omega Jones, a streamer who uses the handle @CriticalBard.

He shared a video snippet on Twitter that showed the chat forum on his screen filling with racist insults as he made his way through a game.

“When marginalised creators on this platform, specifically people of colour, tell you that Twitch needs to do better, just listen,” Jones said.

A Twitch streamer whose profile identified her as bi-racial tweeted that all her friends were being targeted with hate raids, and called for the platform to be made safer.

“We’re launching channel-level ban-evasion detection and account verification improvements later this year,” Twitch said, adding that it is reaching out to streamers to learn more about abuses.