diff options
Diffstat (limited to '')
-rw-r--r-- | emacsconf-publish.el | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/emacsconf-publish.el b/emacsconf-publish.el index 9f83e85..5e15d02 100644 --- a/emacsconf-publish.el +++ b/emacsconf-publish.el @@ -844,10 +844,11 @@ Entries are sorted chronologically, with different tracks interleaved." (length (assoc-default "WAITING_FOR_PREREC" by-status)) (emacsconf-sum :time (assoc-default "WAITING_FOR_PREREC" by-status)) (length talks)) - (let ((list (append (assoc-default "TO_PROCESS" by-status) - (assoc-default "PROCESSING" by-status) - (assoc-default "TO_AUTOCAP" by-status) - (assoc-default "TO_ASSIGN" by-status)))) + (let ((list (append + (assoc-default "TO_ASSIGN" by-status) + (assoc-default "TO_PROCESS" by-status) + (assoc-default "PROCESSING" by-status) + (assoc-default "TO_AUTOCAP" by-status)))) (format "<h1>%s talk(s) to be captioned (%d minutes)</h1><p>You can e-mail <a href=\"mailto:sacha@sachachua.com\">sacha@sachachua.com</a> to call dibs on editing the captions for one of these talks. This year, we're experimenting with using OpenAI Whisper to provide auto-generated VTT that you can use as a starting point. If you're writing them from scratch, you can choose to include timing information, or we can probably figure them out afterwards with a forced alignment tool. Also, if you feel like making chapter markers, that's cool too. More info: <a href=\"https://emacsconf.org/captioning/\">captioning tips</a></p><ul class=\"videos\">%s</ul>" (length list) (emacsconf-sum :video-time list) |