Product lead Kayvon Beykpour said in a tweet, “Really appreciate the feedback we heard from customers on missing accessibility support for our iOS voice experiment yesterday. Sorry we let you down. We’ve already fixed a few issues today (will be in the next release), and exploring more improvements (like transcription) now.”

He also linked to a series of tweets from Twitter Support that provided more details.

Twitter Support said issues related to vision accessibility were addressed, including making voice tweets identifiable on the timeline and other accessibility improvements, and these will be added to an upcoming iOS release.

The social network is also exploring ideas on how to support manual and auto transcription, as well as how to build a dedicated group to focus on accessibility, tooling and advocacy across all products, in partnership with the @TwitterA11y and @TwitterAble teams.

And a passionate, apologetic response came from Twitter product designer Maya Patterson, sent as a voice tweet and written out.

Highlights include: “Point blank period, we fucked up, I fucked up. 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 … As a Black woman, I wholeheartedly understand what it feels like to be excluded, to be the one who has to advocate for myself and my community. And it is frustrating. It is gut wrenching. And it’s not OK. And I’m so sorry … that this feature was probably triggering and honestly a slap in the face to all of the work, the decades of work that you all have been putting in to raise awareness about … including accessible features in your design and development process from the start.”

The main Twitter account followed suit, as well, sharing a Juneteenth message as both a voice tweet and in text form.

Twitter began testing voice tweets Wednesday.

People with access to the feature can open the tweet composer on their iOS device, tap the new icon depicting wavelengths and tap the record button to begin recording.

Voice tweets can be up to 140 seconds in length, and if the time limit is exceeded, a new voice tweet automatically begins, creating a thread.

Tapping the done button will bring users back to the composer screen, where they can share their voice tweets. They will appear in Twitter’s timeline alongside other tweets. Tapping the image will enable people to play the audio.

Playback starts in a new window, so users can listen to voice tweets while continuing to scroll through their timelines.

Voice tweets are available for consumption by everyone on Twitter, whether or not they have access to the creation process.