Cleanvoice
Creative
podcast audio editing
SHOW HN2021
How HN took it
9 positive, 20 neutral or asking, 1 mixed, 14 critical, out of 44 comments about the project. Thread score -0.15.
Idea: 3 for, 5 againstDesign: 1 for, 3 againstExecution: 1 for, 2 againstPrivacy: 0 for, 3 againstDocs: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Useful for podcasters who edit a lot
From 2 comments
As someone who has personally edited over a hundred 1-2 hour podcasts with a new guest every time removing umms, ahhs, dead air and filler words is soul crushing. It has gotten…
nickjj on HNThis is excellent, well done! I'd be curious to know how it's done, as I don't know much about deep learning and this looks like magic to me.
stavros on HN
Impressive ML work
From 2 comments
This is excellent, well done! I'd be curious to know how it's done, as I don't know much about deep learning and this looks like magic to me.
stavros on HNReally cool project, I wish you great success! Could be useful for my (german) podcast agency! Out of curiosity: Which ai-technology did you use? OpenAI? Google API? Or did you…
mikepechadotcom on HN
Objections
Edited audio sounds unnatural with overly long or hard-cut pauses
From 2 comments
As someone who has personally edited over a hundred 1-2 hour podcasts with a new guest every time removing umms, ahhs, dead air and filler words is soul crushing. It has gotten…
nickjj on HNIs the example in the page really made by the computer? In my opinion the pauses in where the filler words were are slightly too long. Is it possible to configure this? Is it…
gus_massa on HN
Terms of service grant overly broad content rights
From 1 comment
The Terms of service seem worrisome. > By posting your Contributions to any part of the Site or making Contributions accessible to the Site by linking your account from the Site…
daenney on HN
Demos do not play on iOS Safari
From 1 comment
Your demos don’t play on iOS safari.
geuis on HN
Requests
Export to editor project files such as EDL/ADL
From 2 comments
This is awesome. Can I suggest the ability to export as project files for popular editors for your roadmap? It'd cut professional workflows down substantially, which would be…
eganist on HNTo add to this, it might also be a good feature to output EDL files for video editors. https://www.rev.com/blog/how-to-import-an-edit-decision-list-edl-into-davinci-resolve This…
tim-- on HN
API access and bulk pricing
From 1 comment
Hey! Justin (from Transistor.fm) here. This looks really interesting. Two questions: 1. Any plans for an API and bulk pricing? 2. Any plans to add loudness normalization,…
mijustin on HN
Video support
From 1 comment
Can anyone recommend similar for removing ums etc. in videos? IIRC there is a workflow in some professional software, but being able to train and throw the algorithim right at…
xipho on HN
Questions
How the models work and approach used
From 2 comments
Without giving away your secret sauce, what are your approaches to the cleaning process? Is it a combination of different passes of algos or is it something more generic and…
qmmmur on HNWhat's the high level approach required to build something like this yourself? Does it involve relying on speech to text with timestamps and then a series of cuts based on that?
fareesh on HN
Latency for near-live use
From 1 comment
How is the latency and, if it's sufficiently low, could this realistically be applied to "nearly live" content? That scenario seems really appealing for conferences, even if it…
nmstoker on HN
Alternatives named
Descript
From 2 comments
Sounds similar to Descript https://www.descript.com/
monroewalker on HNI'm going to sound like a negative nancey, but I wish podcasters/youtubers would just practice their speaking skills instead of rely on series of really quick jump cuts. Worst…
spicybright on HN
Auphonic
From 1 comment
Reminds me of https://auphonic.com/ Their pricing is also similar, but Auphonic allows both subscription and prepaid "credits".
hs86 on HN
Krisp, Overcast, revoldiv
From 3 comments
It would be a huge engineering endeavour, which I wouldn't be capable of doing. That said, things like background noise and some sounds can be removed. See Krisp.ai
autoencoders on HNOvercast has features to do some of this on the listener side. I prefer having the AI on the listener side so I can go back to the raw version if the AI messes up for some reason.
throwaway1777 on HNrevoldiv.com has a similar feature set
userhacker on HN