From bd39eb0278ffaa420543f5817366d9b04d74e5b1 Mon Sep 17 00:00:00 2001 From: GitHub Action Date: Thu, 3 Oct 2024 06:29:41 +0000 Subject: [PATCH] Update Generated Markdown Files --- docs/rss.xml | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/docs/rss.xml b/docs/rss.xml index fbe91fed6d..3499d82318 100644 --- a/docs/rss.xml +++ b/docs/rss.xml @@ -4,9 +4,9 @@ https://fcp.cafe FCP Cafe - Thu, 03 Oct 2024 02:56:25 GMT + Thu, 03 Oct 2024 06:29:41 GMT - Thu, 03 Oct 2024 02:56:25 GMT + Thu, 03 Oct 2024 06:29:41 GMT <![CDATA[23rd May 2023]]> Iain Anderson has written a very detailed Final Cut Pro 10.6.6 blog post over at Pro Video Coalition.

@@ -7126,6 +7126,12 @@ The amazing George Edmondson explained to me that as he discusses in his tutoria

In this case mCaptions is using Whisper.cpp's Large V3 model.

Our first test was with some Strada test clips (that have been provided to Strada beta testers), in Mandarin, and it failed miserably.

+

!!!danger UPDATE (20241003 1630) +MotionVFX has since contacted me and have said:

+
+

As for the issue with analyzing Chinese - we're aware of it and an update should be ready within the next few days. +!!!

+

Next we tried some other Strada test clips, but they had multiple audio channels, and mCaptions wouldn't even import them:

It would be great if mCaptions was smart enough to read the enabled channels from the FCPXML and only use the enabled channels for transcription. Maybe in a future update.