Feature Requests

There are various ways to achieve this goal, so I'll focus on explaining the main concept and capabilities. It's crucial to keep this separate from the 'Filler Words Tool,' and I'll elaborate on the reasons later. Before delving into the tool idea, I'd like to provide some context for my proposal. It's all one paragraph so you can skip it. I'm part of a team dedicated to safeguarding kids from potential threats posed by adults on the internet. Our approach involves setting up decoys, engaging with adults, and conducting thorough interviews as part of our investigations, that can last from 1 to 3 hours. Building rapport is our initial strategy to create a situation where the individuals believe law enforcement won't be involved, which makes confessions easy for them to talk about in depth. We record this on video as evidence for law enforcement, ensuring that any inappropriate online activities are disclosed. However, the challenge arises when law enforcement requires comprehensive information before confiscating devices or making arrests that will prevent an individual from tampering with evidence. This process of censoring 1-hour videos can extend to 4-8 hours to guarantee nothing is overlooked since youtubes guidelines are applied selectively. The proposed tool is essentially a feature to create and manage a list of words or phrases to be muted. If automating this process proves too complex, an alternative is to streamline the manual process. Users could simply right-click and select 'mute word' or 'mute words' (if multiple are highlighted). It's important to emphasize the distinction between 'word' and 'words' for clarity. To optimize loading time, this feature should be an option during media import. Activating it would expand the choice between a default list and a customized one. Users can also decide if they want the words to be highlighted and/or muted. Default lists would include words generally considered inappropriate, avoiding explicit mention to maintain context. Users adding words to the default list would automatically create a separate list comprising default words and any additions. This caters to the most common censored words on TV and addresses monetization limitations on platforms like YouTube. For user privacy, a content warning when viewing the list could be implemented. This ensures users are aware that the list may contain offensive words. Keeping this tool separate from the 'Filler Words Tool' allows for a toggle option on the timeline. When activated, users can visualize all muted sections or highlight a specific area to mute, correcting any oversights or potential misinterpretations. This information can be used to train the AI, enhancing its accuracy over time. Lastly, when users highlight and right-click any text, the menu should include an option to add it to the list. If a single word is chosen, it's treated as one entry. However, if multiple words are selected, they should also be considered as a single entry to account for context. For instance, platforms like YouTube consider context when assessing descriptive sentences—explicit details might be overlooked, but any indication involving a minor can cause a video to reach less people. Our goal isn't for entertainment purposes. It is essential for spreading awareness to parents who can find other resources to keep better eyes on their Childs online activities and even take preventative measures. Not only that, but the more these adults are caught, the easier it is to take down larger groups that actively produce the material being plagued online - leading to the rescue of REAL victims all over the world. So it's more than just a feature request in my eyes. Thanks for anyone that actually reads this. Sorry for it being so detailed. I just think many haven't gave enough thought into the actual implication of such a feature. Devs work hard and it's not as simple as many think. Ive made far more in depth mockups for InShot and PicsArt who both made the ideas into real tools. I wouldn't mind doing that as well.

adding a minimum indentation parameter when exporting. that is, so that the subtitles are interrupted only between paragraphs, but between phrases, some phrase is constantly on the screen, namely the previous phrase until the beginning of the next one

Would be helpful to have the option to not have partial words included in transcripts (sus- suspicious; ev- even) and to have the option to not have quotation marks added when something is quoted, as well as an option to change slang to the proper terms (gonna, going to; wanna, want to; 'em, them)

suggest a domain name. a domain

I don't know what happened recently but for some time, the transcript accuracy really improved but recently, it's quite bad again so I'd like to ask to rollback whatever changes you've made to it.

Support persian language for video editing, translations, transcription, etc.

The German language transcription is completely unusable. Almost all short words such as yes, so, and also ahms, etc. are not recognized. Even entire sentence dividers are not translated when you start again. For a tool that offers transcript text-based editing, this is not a no-go. Unfortunately, the upload of external transcripts via the transcript.srt upload also no longer works??? As a result, even the detour via functioning transcription engines such as Whisper is no longer possible. All in all, Descript cannot be used if the absolute basis (the transcript) does not work properly. Too bad, there's a lot of wind for nothing.

I do not know how to program an AI transcription engine, and that is all I utilize Descript for. First of all, I must add that I have tried other engines available, and only one gets close enough to the performance you all provide - yet, the complexity, required training, required hardware/software, and expense is relatively unreasonable compared to what you offer. The only feature they offer is a comprehensive (yet hard to train) editing tool, which with care, experience, and lots of customization, I have nearly replicated just with MS Word. On the positive side, which is all I have today (though I will try to come up with something to request...), whatever the Descript team has done to the transcription engine lately is SUPERB. It is so good that I have to go to my editing macros and readjust them, because the usual corrections they seek no longer exist. Surprisingly, I see phrases being quoted, and in a 600-page document, with maybe 50 quoted phrases and sentences provided by the AI, only about seven of them were misquoted - and it was mostly because the speakers broke the sentence in ways that the AI could not reconcile. I was surprised and delighted - I did not think this was possible. Additionally, names of weird places that I usually have to hazard a guess on a search engine are easily found and properly spelled by Descript a lot better than before. As what I transcribe is legal, I found even case law properly transcribed - a new one for any AI I have worked with; it is not all-encompassing and nowhere near all correct, but positively surprising nonetheless. This is where the AI gets it wrong or not at all: days and years. For some reason, twenty twenty six does not translate as 2026; one May twenty four does not translate as 1 May 24; et cetera. I have macros for it, but thought you would like to know. That one seems easier to fix at a glance than proper quotations! There are many other numerical issues related to translation, but dates seem like the more standardized I can think of that would benefit the most people. I have also noticed positive differences in the capitalization used by Descript; it catches things that I occasionally forget to capitalize. Military acronyms are getting better formatted as well, which is a very huge surprise... (if you all need military court transcripts and examples, I could get you sanitized versions for your AI dictionary and learning model, as well as lists of weird terms and how they are spoken - I used to teach this material). So, good feedback thus far, and thanks for all the work you all do. Oh, I remember trying out the feature that will replace certain things in the translated text for whatever we gave it... I have no idea if it has been worked on since I stopped using it, but when I tried it (only when it first came out), I immediately saw the value in it but that it needed lots of work. I do not know if you all do beta testing of features with enterprising users that can give you detailed feedback to help you, but I volunteer as long as time allows. :-) Thank you all for the work you do. Thanks to Descript, my productivity is at least 250% more AND better than before using this AI tool, which allows me to work fewer hours for the same productivity, or increase my product (but being a 100% disabled vet that needs constant breaks, I need tools that help me increase hourly productivity, since working more hours is occasionally impossible due to health./physical issues). Again, thanks!! ( I would NOT MIND military discount, even if only on the annual pricing... just saying... LoL)

Every 2-3 hour workshop leaves hundreds of grey words that aren't transcribed chopped in half, meaning I still have to manually review and fix every cut which takes 1-3 hours per long video. Will need to move off descript to timebolt if this is not fixed ASAP.

A long-time 'base' feature of Descript has been to remove filler words. This shouldn't be an AI charge, especially as it appears to essentially just be 'find and replace' or simple search. We've already paid for the transcription and this was an 'easy win' feature for recommending Descript. Please stop charging against AI credits for what used to be an included feature.