VideoHelp Forum
+ Reply to Thread
Results 1 to 13 of 13
Thread
  1. Member
    Join Date: Mar 2014
    Location: US
    Search Comp PM
    I lopped off a good section of an M4A, then added in some labels, creating ~20 tracks out of the one master file. I exported as M4A at the highest quality setting, and the resulting group of files is ~30% larger than the original master file.

    The original file was ~192kbps, and the new files are ~250-300kbps. I obviously didn't get additional quality added in, the export settings I used just made for bigger files. So what could I have done to get all of the original quality without creating larger files?
    Quote Quote  
  2. Member
    Join Date: Sep 2007
    Location: Canada
    Search Comp PM
    It's not possible to get original quality, even at the same bitrate with lossy codecs because you are re-encoding . Even at larger file sizes, the quality has dropped

    re-encoding means it's decoded to uncompressed, then re-compressed . When it's re-compressed, bits of data are thrown away

    mp3directcut can cut/edit aac without re-encoding if all you are doing is trimming sections
    Quote Quote  
  3. Member
    Join Date: Mar 2014
    Location: US
    Search Comp PM
    Originally Posted by poisondeathray View Post
    It's not possible to get original quality, even at the same bitrate with lossy codecs because you are re-encoding . Even at larger file sizes, the quality has dropped

    re-encoding means it's decoded to uncompressed, then re-compressed . When it's re-compressed, bits of data are thrown away

    mp3directcut can cut/edit aac without re-encoding if all you are doing is trimming sections
    Okay. I wonder if mp3DirectCut can create separate tracks out of one master track. Also, will original quality be retained in Audacity if exported as WAV (albeit at a much larger filesize)?
    Quote Quote  
  4. Member
    Join Date: Sep 2007
    Location: Canada
    Search Comp PM
    I don't understand what you mean by "separate tracks" ? You can copy out sections

    WAV is uncompressed, so quality is the same as original. The quality loss occurs when you use lossy compression. MP3, AAC are lossy compression formats. Lossy means you throw away bits of data, even at higher than original bitrates.

    But there would be no quality loss if you used Lossless compression either (such as FLAC) .
    Quote Quote  
  5. vanished El Heggunte's Avatar
    Join Date: Jun 2009
    Location: Misplaced Childhood
    Search Comp PM
    Originally Posted by bqi View Post
    .................

    Okay. I wonder if mp3DirectCut can create separate tracks out of one master track.
    Yes and no First, you'll have a lot of manual work. Secondly,
    lossy audio codecs are not «sample accurate», if your intention is trim the encoded files
    So you'd better go the uncompressed/lossless route in this case.
    Quote Quote  
  6. Member hech54's Avatar
    Join Date: Jul 2001
    Location: Yank in Europe
    Search Comp PM
    I'd go M4A to WAV 44.1, then create chapters(or individual songs/tracks) with CDWave.
    Quote Quote  
  7. Originally Posted by bqi View Post
    Okay. I wonder if mp3DirectCut can create separate tracks out of one master track. Also, will original quality be retained in Audacity if exported as WAV (albeit at a much larger filesize)?
    Yes it can. You may end up with a little "glitch" at the edit point if you're deleting a section of the audio or joining different sections etc. If so, try adjusting it a little. I don't work with AAC audio that way much but I've used mp3directcut for editing MP3s a lot and it has the potential to cause an audible "glitch" at the edit point, so I assume AAC would be the same. If the area where you're editing is silent though, it shouldn't be an issue.

    You can have multiple instances of mp3DirectCut running at a time and you can cut, copy and paste between them. I use that method a bit for MP3. Copying a section of encoded silence from one and paste it into another, or for copying sections from multiple audio files and pasting them into another instance of mp3directcut to create one large file.... that sort of thing.
    Quote Quote  
  8. Member
    Join Date: Mar 2014
    Location: US
    Search Comp PM
    Originally Posted by hello_hello View Post
    Originally Posted by bqi View Post
    Okay. I wonder if mp3DirectCut can create separate tracks out of one master track. Also, will original quality be retained in Audacity if exported as WAV (albeit at a much larger filesize)?
    Yes it can. You may end up with a little "glitch" at the edit point if you're deleting a section of the audio or joining different sections etc. If so, try adjusting it a little. I don't work with AAC audio that way much but I've used mp3directcut for editing MP3s a lot and it has the potential to cause an audible "glitch" at the edit point, so I assume AAC would be the same. If the area where you're editing is silent though, it shouldn't be an issue.

    You can have multiple instances of mp3DirectCut running at a time and you can cut, copy and paste between them. I use that method a bit for MP3. Copying a section of encoded silence from one and paste it into another, or for copying sections from multiple audio files and pasting them into another instance of mp3directcut to create one large file.... that sort of thing.
    That's good to know, thanks. Apologies if this has already been answered, but creating multi-sections out of one file in mp3directcut, is quality lost?
    Quote Quote  
  9. Originally Posted by bqi View Post
    That's good to know, thanks. Apologies if this has already been answered, but creating multi-sections out of one file in mp3directcut, is quality lost?
    No. It doesn't re-encode.
    Quote Quote  
  10. If you're spliting a file into sections you can just highlight the part you want to save and use the "File/Save Selection" menu to save it as a new file. You shouldn't end up with glitches that way. It's usually only when deleting a section from the middle of an audio file and keeping what's left that it might happen, and as I said I've really only worked with MP3s that way so for AAC it mightn't be as much of an issue. I don't really know.
    Last edited by hello_hello; 21st Apr 2014 at 05:16.
    Quote Quote  
  11. Also, will original quality be retained in Audacity if exported as WAV (albeit at a much larger filesize)?
    I missed that part of your earlier question. Wave is a lossless format so the short answer is yes, the original quality will be retained. You won't lose further quality until the wave is re-encoded using a lossy encoder, whether it be AAC or MP3 etc. The long answer......

    Wave is lossless, in that it doesn't throw information away. However there's a limit to how accurately a wave file can store information. Think of it this way.....
    A 24 bit wave file can assign 16,777,216 different values to any particular sample (2 to the power of 24), so it can replicate a waveform very accurately. A 16 bit wave file can assign 65,536 different values, and for 8 bit it's 256 different values. So even though no data is thrown away, if you convert a 24 bit wave file to a 16 bit wave file then information is "rounded" on the way through. Lossy formats don't have a fixed bitdepth, so when converting to wave a similar "rounding" takes place. The higher the bitdepth, the greater the accuracy (and the larger the file size).
    All that's techno-babble to a certain extent though. 16 bit is considered enough to reproduce a waveform accurately which is why the short answer to your question is converting lossy audio to a wave file would be considered lossless.
    I'm fairly sure Audacity applies dithering when converting to wave (someone who uses it may know for sure). Dithering is the introduction of noise when converting in order to randomise the "rounding" which takes place. It's not audible, nothing to worry about, and is commonly used when converting both audio and video. I just thought I'd answer your question thoroughly.
    Quote Quote  
  12. Member
    Join Date: Mar 2014
    Location: US
    Search Comp PM
    That's really good information.

    Originally Posted by hello_hello View Post
    Also, will original quality be retained in Audacity if exported as WAV (albeit at a much larger filesize)?
    I missed that part of your earlier question. Wave is a lossless format so the short answer is yes, the original quality will be retained. You won't lose further quality until the wave is re-encoded using a lossy encoder, whether it be AAC or MP3 etc. The long answer......

    Wave is lossless, in that it doesn't throw information away. However there's a limit to how accurately a wave file can store information. Think of it this way.....
    A 24 bit wave file can assign 16,777,216 different values to any particular sample (2 to the power of 24), so it can replicate a waveform very accurately. A 16 bit wave file can assign 65,536 different values, and for 8 bit it's 256 different values. So even though no data is thrown away, if you convert a 24 bit wave file to a 16 bit wave file then information is "rounded" on the way through. Lossy formats don't have a fixed bitdepth, so when converting to wave a similar "rounding" takes place. The higher the bitdepth, the greater the accuracy (and the larger the file size).
    All that's techno-babble to a certain extent though. 16 bit is considered enough to reproduce a waveform accurately which is why the short answer to your question is converting lossy audio to a wave file would be considered lossless.
    I'm fairly sure Audacity applies dithering when converting to wave (someone who uses it may know for sure). Dithering is the introduction of noise when converting in order to randomise the "rounding" which takes place. It's not audible, nothing to worry about, and is commonly used when converting both audio and video. I just thought I'd answer your question thoroughly.
    Quote Quote  
  13. Member Cornucopia's Avatar
    Join Date: Oct 2001
    Location: Deep in the Heart of Texas
    Search Comp PM
    Couple of caveats & extra suggestion about the above:

    1. Wave LPCM or a lossless encoder will accurately return a copy of your INPUT, whatever that was. If the input was already lossy, the loss has already occurred and you can't get it back. WAV won't get you back to the ORIGINAL original.

    2. There is loss (or rather CHANGE) to the file when processing: SampleRate conversion, Normalization, Noise Reduction, etc...and Dithering. For certain kinds of material, even 1/3LSB dither is noticeable, though rarely unpleasant. Mostly, it is so small of a difference as to be not noticeable to the great majority of listeners.

    3. You can also make playlists, using metafiles, such as m3u, asx/wvx, pls, wpl, smil (assuming your player can work with them). They make it easy to sort of concatenate totally separate items to make for a NEARLY seamless compendium without going through the effort of decoding/re-encoding, or even demuxing/remuxing. Though you seem to be doing the opposite (breaking large work up into "chapters"), you can still make use of same by using the playlist to "play" the separate (now distinct) sections of the master file, all without having to reencode. SMIL's xml "chaptering" is fully supported by QT/MOV, so I'm guessing it got passed along and is also supported by MP4/M4a as well (being that they were derived from QT).

    Scott
    "When will the rhetorical questions end?!" - George Carlin
    Quote Quote