[Libav-user] av_rdft_calc() resulting buffer

Ricky Huang rhuang.work at gmail.com
Sat Jun 7 01:54:37 CEST 2014


On Jun 4, 2014, at 6:09 PM, Ricky Huang <rhuang.work at gmail.com> wrote:

> Hello all,
> 
> According to a LAS blog post here: http://blog.pkh.me/p/6-las-lossy-audio-spotter.html.  It says that "the buffer is composed of couples of real and imaginary floats except for the two first ones which are the real parts of the first and last bins…" under the "Processing the audio data" section.
> 
> Does anyone know where that is documented?  I looked on ffmpeg's doc (http://www.ffmpeg.org/doxygen/2.2/group__lavc__fft.html#ga745b8757713a5a0f0347eddcc6ccc9a4) and sees nothing on that.  I also looked at the avf_showspectrum.c code and it doesn't seem to make the special case for those two positions.

Clément, the author of the LAS article, answered the question for me.  It is located in a comment in libavcodec/rdft.c:

   /* i=0 is a special case because of packing, the DC term is real, so we
      are going to throw the N/2 term (also real) in with it. */

and can be found here: http://www.ffmpeg.org/doxygen/2.2/rdft_8c_source.html#l00071

Just in case anyone run into the same question in the future.


> 
> 
> Thanks in advance.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://ffmpeg.org/pipermail/libav-user/attachments/20140606/0b5abe02/attachment.html>


More information about the Libav-user mailing list