flac dev - Mar 2012 - Predicting the loss of data from flacs to mp3

Hello,

My partner and I are both high school students who are working on a
device that encrypts secret messages into sound files. The method that
we are using involves the conversion of data into an mp3 format, and
this results in a loss of data that is crucial to the success of the
device. We would like to be able to predict which points are going to
be modified significantly or lost completely when we write our data as
an mp3. Could you help us?

We were able to achieve success with .flac files (since they're a
lossless audio format), but with mp3 files, were unable to recover the
secret message since the mp3 compression altered the data points. We
were thinking that if we could know which points (in the time domain)
would be adversely affected, we could avoid encrypting in those points. Any
clarification on how .mp3 files compress data would be very helpful.
Thank you!

Hi,

Predicting loss of data when encoding mp3 is more suitable for an mp3
discussion forum instead of a lossless format one. A pointer - it isn't
possible. It heavily depends on the encoder, compression quality and
specific settings. No one will be able to give you a definite answer, as
the psychoacoustic model is a complex thing. The best thing you can do is
encode the data in mp3 and then compare it. If you really need to predict
it - transients ( http://en.wikipedia.org/wiki/Transient_(acoustics) ) are
the most difficult to encode with mp3.

Regards,
Ivailo Karamanolev

On Tue, Mar 20, 2012 at 1:27 PM, Ajay Mandlekar <ajay.mandlekar at
gmail.com>wrote:
> Hello,
>
> My partner and I are both high school students who are working on a
> device that encrypts secret messages into sound files. The method that
> we are using involves the conversion of data into an mp3 format, and
> this results in a loss of data that is crucial to the success of the
> device. We would like to be able to predict which points are going to
> be modified significantly or lost completely when we write our data as
> an mp3. Could you help us?
>
> We were able to achieve success with .flac files (since they're a
> lossless audio format), but with mp3 files, were unable to recover the
> secret message since the mp3 compression altered the data points. We
> were thinking that if we could know which points (in the time domain)
> would be adversely affected, we could avoid encrypting in those points. Any
> clarification on how .mp3 files compress data would be very helpful.
> Thank you!
> _______________________________________________
> flac-dev mailing list
> flac-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/flac-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/flac-dev/attachments/20120921/1d76d059/attachment.htm