Questions from listeners

Listeners can send the speaker a question in their own language, from the same page they listen on. The speaker reads it in the language they speak and answers out loud, and the answer is interpreted into every language like the rest of the talk. No one needs an app or an account.

It's off until the organizer turns it on. It's part of Pro (and Max and Enterprise) and costs nothing beyond the plan. On Basic the card shows the switches turned off, with a link to the plans. If Pro ends, listeners stop seeing the option; questions already sent stay on the event page.

Turning questions onAsk

On the event page, the Questions from listeners card has two switches. Turn on either or both:

• Typed questions: listeners write a question.
• Spoken questions: listeners record a question of up to a minute.

Listeners see the option as soon as a switch is on, also mid-event. Turning both off hides it again; questions already sent stay on the event page.

How listeners askAsk

1. Under the player, tap Ask the speaker a question.
2. Optionally type a name. Without one, the question is anonymous.
3. Type the question and press Send, or press Record a question, speak, and press Stop and send. A recording stops by itself after 60 seconds.

After sending, the listener sees the question as the speaker will get it in their own words: for a spoken question, the text it was turned into, so they can tell whether it was heard right. A recording shorter than a second isn't sent.

The listener's language is the language they are listening in. Someone on the original audio asks in the spoken language.

Recording doesn't start. The browser asks for the microphone the first time; the listener must allow it. If it was blocked, allow the microphone for the site in the browser's settings and reload. Spoken questions need a browser that can record audio; on very old browsers only typing is offered.

"You have sent many questions". One listener can send 6 questions an hour, and one network 30, so a single person can't flood the speaker. The limit resets within the hour.

Reading and answering questionsAsk

Questions appear on the event page under Questions from listeners, and on the speaker screen at the top of the side panel, so the speaker can take them while on air. Both update by themselves every few seconds.

• Each question shows in the room's spoken language in large text, with the listener's own words, their language, their name if given, and the time beneath. A microphone icon marks a spoken question.
• They come oldest first, so they can be taken in turn.
• In an event with several rooms, a question goes to the room the listener was in. On the speaker screen each room sees its own questions and those asked of the event as a whole; the event page shows which room each one came from.
• Press Answered once it's dealt with, or the Dismiss (×) button to set it aside. Either hides it from the list. Show answered too shows them again, and Reopen puts one back.

Anyone who can run the event can read and close questions, the speaker included. Listeners don't see each other's questions.

TranslationAsk

Questions are translated by AI into the spoken language of the listener's room (for a question to the event as a whole, the first room's). A question already in the speaker's language isn't translated. If translation fails, the question still arrives in the listener's own words.

Like the interpretation, translation can make mistakes. When a question matters, the original is right under it. Spoken questions are turned into text by AI too; with background noise or a quiet voice, the text can be off.

PrivacyAsk

• The organizer and the people running the event see the questions, with the name if the listener gave one.
• A spoken question's recording is turned into text and not kept. Only the text is stored.
• Questions are kept with the event and deleted when the event is deleted.
• Questions are processed by OpenAI to translate and transcribe them, as the interpretation is. See Privacy, recordings and security.