One of the designers behind Twitter’s voice tweets has apologised for failing to make the feature accessible to deaf and hard of hearing people.

The social network launched a test of the new option among a sample of users this week, allowing them to record up to 140 seconds of audio to post as a tweet.

But the announcement was met with concerns about it excluding those with hearing disabilities.

Although the firm initially responded saying it was “exploring ways” to make voice tweets accessible, critics argued accessibility should be factored in from the beginning, not at the end.

Maya Patterson, staff product designer behind the project, has said it was a “huge error” and the team is “working to fix this as soon as possible”.

“We f***** up, I f***** up,” she tweeted.

“We launched a test and we should have included accessibility features in that test.

“That was a huge error, and one I personally have made myself sit with all day today.”

Ms Patterson – who joined Twitter less than six months ago – said, as a black woman, she “wholeheartedly” understands what it feels like to be excluded and wants to educate herself fuller on what she can do as a designer to be a “true ally” in her work.

“It is frustrating,” she explained.

“It is gut wrenching. And it’s not okay. And I’m so sorry.”

Voice posts appear in Twitter timelines alongside traditional tweets, represented by an animated video of the poster’s profile picture.

Users will find the feature in the tweet composer screen as a new icon featuring purple wavelengths, which takes users to the record screen to capture their audio.