som student Technickej Univerzity v Kosiciach a pisem bakalarsku pracu na temu Vyhladavanie informacie v audiodokumentoch alebo aj CBID. Moj veduci je doc. Jozef Juhar Csc. mozno ho aj z pocutia poznate on sa viac zameriava na rozpoznavanie reci. Ale aby som presiel k jadru mojej otazky. V BP sa chcem poukazat na musicbrainz.org a softwary ktore ste napisali. Jedna sa mi o detaily prace fingerprintov a system vyhladavania v databaze. Za odpoved vopred dakujem.

S pozdravom

Tomas Koctur
Radio 9

It works fine for me, and it is so simple…

C:\Fuentes\includes\taglib-1.7.1>c:\cmake-2.8.4-win32-x86\bin\cmake.exe -DZLIB_I
NCLUDE_DIR=c:\Fuentes\includes\zlib\include\ -DZLIB_LIBRARY=C:\Fuentes\includes\
zlib\win\static32\zlibstat.lib -DCMAKE_BUILD_TYPE=Release -DWITH_ASF=ON -DWITH_M
P4=ON -DBUILD_EXAMPLES=ON
– Building for: Visual Studio 10
– Check for working C compiler using: Visual Studio 10
– Check for working C compiler using: Visual Studio 10 — works
– Detecting C compiler ABI info
– Detecting C compiler ABI info – done
– Check for working CXX compiler using: Visual Studio 10
– Check for working CXX compiler using: Visual Studio 10 — works
– Detecting CXX compiler ABI info
– Detecting CXX compiler ABI info – done
– Found ZLIB: c:/Fuentes/includes/zlib/include (found version “1.2.5″)
– Ensure you cppunit installed version is at least 1.12.0
– Looking for __RTTI___ in Crun
– Looking for __RTTI___ in Crun – not found
– Configuring done
– Generating done
CMake Warning: The variable, ‘CMAKE_BUILD_TYPE’, specified manually, was not used during the generation.
– Build files have been written to: C:/Fuentes/includes/taglib-1.7.1

Then:

C:\Fuentes\includes\taglib-1.7.1>nmake

Microsoft (R) Program Maintenance Utility Version 10.00.30319.01
Copyright (C) Microsoft Corporation. All rights reserved.

NMAKE : fatal error U1064: MAKEFILE not found and no target specified
Stop.

I tried open the solution generated by cmake, but when compiling it fails creating the library, with this error:

Link:
2> Creating library C:/Fuentes/includes/taglib-1.7.1/taglib/Debug/tagd.lib and object C:/Fuentes/includes/taglib-1.7.1/taglib/Debug/tagd.exp
2>id3v2frame.obj : error LNK2019: unresolved external symbol _uncompress referenced in function “protected: class TagLib::ByteVector __thiscall TagLib::ID3v2::Frame::fieldData(class TagLib::ByteVector const &)const ” (?fieldData@Frame@ID3v2@TagLib@@IBE?AVByteVector@3@ABV43@@Z)
2>C:\Fuentes\includes\taglib-1.7.1\taglib\Debug\tagd.dll : fatal error LNK1120: 1 unresolved externals

thanks for your help!

Do you think this approach would work well to recognize a name of the person from a list of thousands of names? For example, if a person is asked to pronounce his name, store some metrics of that and then, when there is an incoming call and the persons name is spoken, (by a different person) those metrics could be matched? A telephony application? There is some commercial potential here.

Please let me know.

Thanks.

thanks for the wonderful explanation of the algorithm – it’s very clear and easy to understand. I am investigating whether it is possible to implement automatic identification of bird songs from a recording against a DB of known birds.

Judging by the description, and by your feedback to other people, I figure that this algorithm, at least in its current form, may not be suitable for bird songs because of the differences amplitude, frequencies and durations of a song between two different birds (the same specie).

I am wondering though whether there is a modification of the algorithm that might work on bird songs, or whether you are aware of a different approach that might be applicable. Would appreciate any pointers – from general ideas to references to publications.

Thanks.

Denis