Feature Requests

It catches duplicates of, e.g., "it its", but it doesn't catch "a an" and it should

When using "remove retakes", the AI tends to re-construct sentences using multiple takes. It tries to use the first words from one of the first takes, and stitch them with other cuts to complete a take. It should prefer using one clean retake instead of making multiple cuts. Example before: (Multiple cuts stitched together) "Individual pages are attributes. |So in this example, your entity attribute pair is |Ranking factors |And Google search Console." Example after: (Use best take to avoid cuts) "Individual pages are attributes. so in my example, the entity attribute pair is ranking factors and Google search console." When one sentence has 3-4 cuts, it sounds really choppy. There is often at least 1 smooth take that doesn't require so many cuts.

When using "remove retakes", the AI tends to re-construct sentences using multiple takes. It tries to use the first words from one of the first takes, and stitch them with other cuts to complete a take. It should prefer using one clean retake instead of making multiple cuts. Example before: (Multiple cuts stitched together) "Individual pages are attributes. |So in this example, your entity attribute pair is |Ranking factors |And Google search Console." Example after: (Use best take to avoid cuts) "Individual pages are attributes. so in my example, the entity attribute pair is ranking factors and Google search console." When one sentence has 3-4 cuts, it sounds really choppy. There is often at least 1 smooth take that doesn't require so many cuts.

The Underload feauture should be able to keep a history of prompts that were inputed in the past, so that when users come back to the project, they remember where they left off

Hello, I'm never able to use the "gaps shortener" because it always (at least 50% of the time) cuts out 0.5s or 1s of the following audio... so if there is an audio, then silence and then audio... the second audio begins with the first word cut in half. This doesn't allow me to use that feature... moreover, the "uhm" detection gives me back 0 results but then you can clear see a bunch of "uhm" in the text... I suppose it's due to my italian language but can I hope in some fixes? Thanks, Amir

The shorten word gaps feature got bad after the update. Grays out parts that speaker was actually talking and includes those in the cuts which makes the whole thing useless. I can't use it to wipe all the word gaps like before. Can you please fix it?

Be nice to have in the AI Tools a quick way to just generate some keywords to be used in YouTube, etc. I can do this now by typing out my prompt in Underlord but having it there while I'm generating my show notes, etc would be helpful (or make it part of the generated show notes/YT description.

What is it? and how does that work? 1 credit how much is on time?

PLease add the ability for Underlord to use user generated layouts. Underlord is also incredibly confident is does things it absolutely did not do. It hallucinates probably more than any other LLM