Skip to content

james unna

My feedback

1 result found

  1. 217 votes
    Vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
    You have left! (?) (thinking…)
    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
    An error occurred while saving the comment
    james unna commented  · 

    I would happily pay for an add on from an app I like so much. The feature I would like is for shell file to be left in place of duplicate files (if desired) when, for example, the same song is encoded as:
    ~ a 10mb, high quality vbr mp3, (average bit rate 256kb/s) on an artist's album
    ~ a 20mb lossless .flac file on a "best of" album
    ~ a 5mb 192kb/s .aac on a various artists compilation file

    I can choose one or more versions to keep and delete the rest while still leaving album folders untouched so if I decide to copy an album to my laptop where a shell file has been left the shell will refer the copy request to the file in whatever album the "real" audio track has been left.

    To choose the best choice to keep for me I would like to be able to listen to key parts of the track "side by side" so to speak, 5 seconds of a, 5 seconds of version b, and like at the optician you say "b sounds better" then the same with "b.. or c"...

    For those who aren't sure, or who know their children might have a different opinion with their better ears, the "Most popular option". For real technical types, details on where the audio differs, for example: frequency range, bit rate by frequency range, lossiness of encoding... even AI analysis of likely version closest to original or with rules like "accept lossy encoding where: .mp3 compression is to no less than 192kb/s or 160kb/s vbr AND file size is smaller than ½ .flac file or ⅒ .wav etc

    Even better would be completely reliable AI that after learning your patterns of choice regarding weight you assign to quality, file compression ratio, and absolute size it then just correctly removed the files it has a high confidence that you would have removed then only seeks confirmation on marginal choices. As it you provide it more and more data it should bet better and better at choosing. A similar approach could be taken towards images and video.

    james unna supported this idea  · 

Feedback and Knowledge Base