Blank acoustic fingerprint when using Chromaprint - c++

I am trying to differentiate between multiple wav files using acoustic fingerprinting. I am using Chromaprint from AcoustID. I am using 32 bit windows. I have downloaded the file fpcalc.exe and am trying to run it on multiple wav files. The problem is, it is generating blank FINGERPRINT for those wav files.
I am running the command
fpcalc.exe -raw <FILENAME>.wav
The files are in wav format and the size of the files is 1 SEC. I am running the same command on bigger files, then the FINGERPRINT is working fine.
Any pointers?

The Chromaprint library only works for larger files, I think the files should have at least 10 seconds to fingerprint.
Maybe you should look for alternatives like Python and Ruby who have libraries to fingerprint small audio files (if you search a bit).
You can use the gem that I made for wav files:
https://rubygems.org/gems/audio-fingerprint

Related

GLib-ERROR : Creating pipes for GWakeup: Too many open files

I am getting this error when running my c++ program under ubuntu 20.04 with clang 9 :
GLib-ERROR : Creating pipes for GWakeup: Too many open files
I know I have a lot of open files, this is a CGI renderer with a mip-map tiled texture cache, it loads only parts of images in the cache when needed and not the full images, thus keeping files open. There are about 3000 images but it opened only about 1000 before crashing.
What I don't understand is that these texture files are read using stdlib fopen(), not GLib/GTK. So why is the error coming from GLib?
The limit on open files is process- or system-wide. So it doesn't matter whether they are opened by GLib or not (which also uses the same operating system calls to open files, anyway). If you have 1000 files opened at the same time with fopen(), consider closing them if you can.

Quazip too many files

I'm having a problem with QuaZip.
I have a program that extracts files from zip archives. The unzip library I was using hadn't been maintained in about 10 years and had problems with the archive contained more than about 65000 files, so I converted my program to use QuaZip.
It turns out that QuaZip has the same problem. I have an archive that 7-Zip shows has 67946 objects. QuaZip only sees 2397 files. ZipInfo also shows only 2397 files. It appears that these tools have not been updated for the newer zip64 format, although QuaZip does seem to use QuaZipFileInfo64 which presumably is there to support the ZIP64 format. Also, Wikipedia documents QuaZip as supporting the new format.
Calling getEntriesCount() for the quazip object returns 2397.
Using the JlCompress::getFileList code to get the list of files only returns 2397.
Not 100 percent sure what I am asking. Does QuaZip support more than 65500 files in an archive? If so, is there a reason why it would be acting as if it doesn't?
As I suspected in my last comment, the number of entries field in the central directory header is incorrect having overflowed on the creation. The solution I came up with was to remove the code in unzGoToNextFile that assumes that that entry is correct.

TIMIT Corpus. Cant play .WAV audiofile but can play it if I convert it into .mp3

I downloaded the files from the timit speech corpus found here: https://github.com/philipperemy/timit
When I try playing the .WAV files with groove music or windows media player I get an error saying the file cannot be played.
However after converting the file to mp3 using an online converter, the file suddenly works.
The .WAV files are only around 100-200KB in size.
Does anyone know what the reason for this problem might be?
Found the answer here: reading a WAV file from TIMIT database in python
Apparently it is not actually a .WAV file but a NIST file.

error regarding ffmpeg while using python to access a wav file

When I used python audio segment to open a .wav file and divide it into many .wav files i am getting this error
"C:\Python27\lib\site-packages\pydub\utils.py:165: RuntimeWarning: Couldn't find
ffmpeg or avconv - defaulting to ffmpeg, but may not work
warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work"
, RuntimeWarning)"
It seems like you don't have ffmpeg, which is listed in the dependencies section of the Pydub GitHub. However, it is only required if you wish to load or save non-WAV files.
This message is just a warning, if you're using Pydub solely for WAV files, it is safe to simply ignore it.

Metadata of avi files

I used the od tool in Linux to look the metadata of mp3 files and avi files by using the command
od -c filename.avi
For mp3 files it is showing all the fields very properly even in sequence. But for avi files it is not showing the metadata fields.
Whether is there is any difference in the way the metadata fields are stored in avi and mp3 files?
Can somebody tell me some good way of extracting metadata fields of avi files.
GNU lib extractor might be good for what you're looking for:
http://www.gnu.org/software/libextractor/