VideoHelp Forum
+ Reply to Thread
Results 1 to 5 of 5
Thread
  1. Hi everyone.
    i have two audio file that one of them restored of my friends that dont give me method.

    this is Original Audio
    https://www.dropbox.com/s/6zpeh30z1bwzjb1/ORIGINAL.mp3
    and its Restored
    https://www.dropbox.com/s/qhuexh2cjyc1xc0/RESTORED.mp3

    can anyone tell me how can i find out it?
    Quote Quote  
  2. Member
    Join Date
    Mar 2008
    Location
    United States
    Search Comp PM
    Caution - those files are not mp3's. I scanned them with A/v but came up negative.
    They look suspicious when viewed in Hex
    Quote Quote  
  3. Originally Posted by davexnet View Post
    Caution - those files are not mp3's. I scanned them with A/v but came up negative.
    They look suspicious when viewed in Hex
    my mean for improve quality
    Quote Quote  
  4. Member
    Join Date
    Mar 2008
    Location
    United States
    Search Comp PM
    Actually I tried downloading again, different files now - at least they play
    Quote Quote  
  5. Any time you "Restore" audio, you are processing it through an algorithm. In order to find out how the audio was processed, you'd need a full lab, and original WAVEFORM duplicates. With MP3, you are reading a "Lossy" format. Basically, a lot of data is thrown out for compression. To get back to a working Waveform, most apps re-interpolate the MP3 into a new file, but they each have their own way of doing this. Without knowing which app and what year\version of that app was used, finding out exactly what was done is going to be very difficult, mainly because apps will WRITE the MP3 file differently, throwing out data in slightly different ways.

    The most common way to restore or enhance audio is with an EQUALIZER plugin or hardware device. These take in the frequency data, and the amplitudes of each frequency at each given sample. While playing the file back, an engineer will make educated guesses about what needs to happen, and make adjustments. In many DAW apps (those that go beyond just record and playback and plugin) an engineer also has the ability to make "Automated" adjustments, which are another "track" that records changes they make for that time slot in the audio. Many engineers will do this by duplicating a file to multiple tracks, listening for the changes in the structure, then marking those areas in an Automation track, so they can apply a plugin to each structure area and have the tracks crossfade together, making the attenuation of the adjustments more seamless. Some will do this so they can adjust each type of sound (ie music vs dialog vs sound effects) and then slowly bring them all back together and attenuate each one accordingly, using automation to control when each one gets louder or softer.
    In some DAW programs, automation can be copied between tracks, so that layers of plugins can be used to make different changes at different times in the file.
    Again, the Equalizer is the most common plugin, but there are many type of Equalizer (EQ). The next type of plugin is the Gate\Compressor\Expander, which focuses only on total amplitude in each area it "handles" and kicks in to adjust the volume of that area based on your settings. The area gets wider with longer attack and release; the attack side being how far ahead of a large amplitude increase the effect will start to kick in, and the Release for how long after it reaches the selected value it will continue to apply. There's also a KNEE function which, when applied to either attack or release, adjusts how harshly the adjustment kicks in. A soft knee, with a longer Attack\Release will slowly adjust from the initial Attack up to the point where the audio crosses the "Threshold", while a hard knee instantly affects the output as soon as the threshold value is detected within the range set by the attack\release.
    Listen to your files carefully. If they are crisp and clear for vocals, and instruments are very clear, there was some work done to the high frequencies above 2k (2000 cycles\hertz). If there is a fuller thud of percussion, and consonant letter sounds in the speech are full, not scratchy, there was probably some bass work in the lows from 60 to 300 cycles. If the instruments sound more orchestrally joined, less separately clear, or multiple vocals sound a little muttled together like a choir, where the words are perfectly clear but clear enough, then the midrange from 400 to 800 was definitely affected. If masculine voices are full and hearty, 200-850 were played with. For feminine, 325 to 950.
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!