Audience Insider will read your briefing out loud. That is not the same as rehearsing it.
The briefing now has a narrated audio version alongside the written report, built as a separate script-and-narration step so a failed recording never costs you the words. Here is what it actually does, and why it is not a substitute for the tool's other voice feature.
Open a finished briefing in Audience Insider now and there is a second way to take it in: a short narrated version you can put on before you drive to the venue. It sits under a "Listen to it" panel, below the written report. Press the button, and in about half a minute you have a script and, if narration succeeds, audio to go with it.
It is a nice feature. It is also easy to mistake for more than it is.
The narration is a rewrite, not a reading
The panel is explicit about this, and it is worth taking at face value: it is not a reading of the report. A separate pass writes a script aimed at four minutes, built to cover what will actually change how you speak rather than repeating every sourced claim and inference the way the written page does. The report is reference material. The script is the same material, cut down to what you would actually want in your ear on the way to a stage.
That distinction is also why the two steps are separate in the code, not one call that does both. Writing the script is one chat request. Turning it into audio is a much longer job, several thousand characters run through text-to-speech. Splitting them means a narration failure does not take the script down with it. You still get the words, with a plain note that the audio did not come through, instead of an error page and nothing to show for the wait.
