- From: Steven Rowat <steven_rowat@sunshine.net>
- Date: Wed, 26 Mar 2025 18:56:28 -0700
- To: Manu Sporny <msporny@digitalbazaar.com>
- Cc: "public-credentials@w3.org" <public-credentials@w3.org>
Received on Thursday, 27 March 2025 01:56:35 UTC
On 2025-03-26 5:53 pm, Manu Sporny wrote: > Thanks for the feedback, it's super helpful. If there are other things that might be good to summarize... or shorter length... or keep it the same length... any feedback would be helpful as we fine tune the new system. My comment/question isn't about the Summary (which is good to have also), but I noticed specifically the difference for the transcript, between the two different types that arrived in my inbox today: the one for the VC API, versus the one for the CCG Credential CG call. I've been in the habit of spending a few minutes, for the CCG Credential transcripts, using my word processor to strip the words 'um' and 'uh' out of the transcript text, because having them there makes the text very hard to decipher at some points. I find I don't want to read it unless I do this. But this VC API transcript has zero 'um' and 'uh'. It has a few minor peculiarities -- like no capitals at the start of sentences -- but otherwise it's very clean. Which is great, and easy to read. So: Two questions: 1) Will you be able to 'flip that switch' for the CCG group transcripts also, to have this cleaner version? 2) If the cleaner version is being produced by AI (I'm just guessing that it is), is there a danger, or have you noticed, any creep of hallucinations? In other words, is the transcript possibly less accurate in some way with the new system? Steven Rowat
Received on Thursday, 27 March 2025 01:56:35 UTC