Importing corrupt MP3 file - no sound

New to Audacity so giving alot of info.

I am trying to import an dictaphone MP3 file that is corrupt. (the file doesn’t open in win media player, adobe audition etc)
The file was deleted/moved from a memory stick and i recovered it using Recuva recovery software.
The file format is MP3 sound format and the size is 99MB - it has no length when displayed in Windows explorer.

When it imports into Audacity its blank, empty i.e. no sound/noise, wavelengths.
Also, the dictaphone file should be over an hour long but is imported with less time e.g. 40 min.

I’ve tried importing the file (not recognised), then importing the raw data - with the raw data i tried a few settings but didn’t know the exact settings to use e.g. Encoding, Byte order.

I’m using Audacity 2.0.5. - the exe, on Windows Vista.

In Audacity both the corrupt and a working dictaphone MP3 file give the following properties:
MME, Speakers, Microphone, 2 (stereo) input channels, 44100 Hz, 32-bit float.


A working dictaphone MP3 file in VLC player gives the following Codec properties.
Codec: MPEG Audio layer 1/2/3 (mpga)
Channels: Stereo
Sample rate: 44100 Hz
Bitrate: 192kb/s.

A working dictaphone MP3 file in Abobe Audition gives the following properties.
Bit Depth: 32 (float)
Source: MP3 192 Kbps CBR

My questions are:

  1. What raw data settings do i use?
  2. If these raw settings do not work what else can i do?


Sadly, I think you are out of luck. :frowning: Importing the raw data isn’t going to work because in that case Audacity is looking for uncompressed PCM data.* The “raw” MP3 data is encoded in a way that it’s not recognizable as audio at all unless it’s properly decoded.

You might try a different file recovery tool.

I don’t have a lot of hope, but here are some tools that can diagnose and fix some MP3 problems:
[u]MP3 Diags[/u]

***** [u]This Tutorial[/u] shows how a digitized waveform is represented by a series of sample values and how you “connect the dots” to reconstruct the waveform.

An uncompressed WAV file is a header (with the sample rate & bit-depth information, etc.) followed by a series of numbers which are the sample values. If you read the bytes and re-assemble the samples correctly, you can create the wave from the raw data. But, an MP3 file does NOT contain a series of sample values and importing it as raw data will give you “random” sample data that sounds like pure noise.

Hi DVDdoug,
Thanks for the reply, info and tutorial link. I tried a number of different recovery software and after your reply a number of MP3 repair tools - no joy. :frowning: