Menu ▾ ▴

Installing Kaldi on Mac OS X 10.9 (Mavericks)

Help
enzyme156
2014-07-15
2014-07-24

  • enzyme156

    enzyme156 - 2014-07-15

    Hello,

    I am having problems installing Kaldi on my Mac OS X Mavericks. I have successfully installed everything required in /tools using tips from this thread: https://sourceforge.net/p/kaldi/mailman/message/31618662/. Now I am moving onto /src. ./configure and make depend worked fine, but make returned this error:

    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
    make[1]: Nothing to be done for all'. /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]: Nothing to be done forall'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
    make[1]: Nothing to be done for `all'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
    g++ -msse -msse2 -Wall -I.. -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -rdynamic -DHAVE_CLAPACK -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o context-dep.o context-dep.cc
    g++: error: unrecognized command line option '-rdynamic'
    make[1]: [context-dep.o] Error 1
    make:     [tree] Error 2

    Editing kaldi.mk with the version suggested for Mac OS X 10.9 from the thread linked above I also got an error:

    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
    make[1]: Nothing to be done for all'. /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]: Nothing to be done forall'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
    make[1]: Nothing to be done for `all'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
    c++ -msse -msse2 -std=c++11 -stdlib=libstdc++ -Wall -I.. -fPIC -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -DHAVE_CLAPACK -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o context-dep.o context-dep.cc
    c++: error: unrecognized command line option '-stdlib=libstdc++'
    make[1]: [context-dep.o] Error 1
    make:     [tree] Error 2

    Removing '-stdlib=libstdc++' just to see what happens gives:

    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
    make[1]: Nothing to be done for all'. /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]: Nothing to be done forall'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
    make[1]: Nothing to be done for `all'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
    c++ -msse -msse2 -Wall -I.. -fPIC -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -DHAVE_CLAPACK -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o context-dep.o context-dep.cc
    In file included from /opt/local/include/gcc48/c++/unordered_map:35:0,
    from ../util/stl-utils.h:32,
    from ../tree/event-map.h:25,
    from ../tree/context-dep.h:22,
    from context-dep.cc:18:
    /opt/local/include/gcc48/c++/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support is currently experimental, and must be enabled with the -std=c++11 or -std=gnu++11 compiler options.
    #error This file requires compiler and library support for the \ ^
    In file included from ../tree/event-map.h:25:0,
    from ../tree/context-dep.h:22,
    from context-dep.cc:18:
    ../util/stl-utils.h:37:12: error: 'std::unordered_map' has not been declared
    using std::unordered_map;
    ^
    ../util/stl-utils.h:38:12: error: 'std::unordered_set' has not been declared
    using std::unordered_set;
    ^
    In file included from /System/Library/Frameworks/vecLib.framework/Headers/vecLib.h:31:0,
    from /System/Library/Frameworks/Accelerate.framework/Headers/Accelerate.h:20,
    from ../matrix/kaldi-blas.h:56,
    from ../matrix/matrix-common.h:25,
    from ../matrix/kaldi-vector.h:24,
    from ../matrix/matrix-lib.h:24,
    from ../tree/context-dep.h:23,
    from context-dep.cc:18:
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt16 vU16HalfMultiply(vUInt16, vUInt16)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84: note: use -flax-vector-conversions to permit conversions between vectors with differing element types or numbers of subparts
    vUInt16 vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84: error: cannot convert 'vUInt16 {aka     vector(8) short unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mullo_epi16(m128i, m128i)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vSInt16 vS16HalfMultiply(vSInt16, vSInt16)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:377:84: error: cannot convert 'vSInt16 {aka     vector(8) short int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mullo_epi16(m128i, m128i)'
    vSInt16     vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vU32FullMulEven(vUInt32, vUInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:432:55: error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_srli_epi64(m128i, int)'
    vbasicops_vA = _mm_srli_epi64(vbasicops_vA, 32);
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:433:55: error: cannot convert 'vUInt32 {aka     vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_srli_epi64(m128i, int)'
    vbasicops_vB = _mm_srli_epi64(vbasicops_vB, 32);
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:434:56: error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(m128i, m128i)'
    return _mm_mul_epu32(vbasicops_vA, vbasicops_vB);
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vU32FullMulOdd(vUInt32, vUInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:459:82: error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(m128i, m128i)'
    vUInt32 vbasicops_vB) { return _mm_mul_epu32(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vU64Sub(vUInt32, vUInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:628:82: error: cannot convert 'vUInt32 {aka     vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_sub_epi64(m128i, m128i)'
    vUInt32     vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vSInt32 vS64Sub(vSInt32, vSInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:698:82: error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_sub_epi64(m128i, m128i)'
    vSInt32 vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vU64Add(vUInt32, vUInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:765:82: error: cannot convert 'vUInt32 {aka     vector(4) unsigned int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_add_epi64(m128i, m128i)'
    vUInt32     vbasicops_vB) { return _mm_add_epi64(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vSInt32 vS64Add(vSInt32, vSInt32)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:830:82: error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_add_epi64(m128i, m128i)'
    vSInt32 vbasicops_vB) { return _mm_add_epi64(vbasicops_vA, vbasicops_vB); }
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vLL64Shift(vUInt32, vUInt8)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:930:91: error: cannot convert 'vUInt8 {aka     vector(16) unsigned char}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_and_si128(m128i, m128i)'
    _mm_and_si128(    vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F )));
    ^
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'vUInt32 vLR64Shift(vUInt32, vUInt8)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:968:91: error: cannot convert 'vUInt8 {aka vector(16) unsigned char}' to 'm128i {aka vector(2) long long int}' for argument '1' to 'm128i _mm_and_si128(m128i, m128i)'
    _mm_and_si128(__vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F )));
    ^
    In file included from ../base/kaldi-common.h:36:0,
    from ../itf/context-dep-itf.h:21,
    from ../tree/context-dep.h:21,
    from context-dep.cc:18:
    ../base/kaldi-math.h: At global scope:
    ../base/kaldi-math.h:84:17: warning: 'kaldi::kBaseLogZero' defined but not used [-Wunused-variable]
    const BaseFloat kBaseLogZero = -std::numeric_limits<BaseFloat>::infinity();
    ^
    ../base/kaldi-math.h:87:17: warning: 'kaldi::kBaseFloatMax' defined but not used [-Wunused-variable]
    const BaseFloat kBaseFloatMax = std::numeric_limits<BaseFloat>::max();
    ^
    make[1]: [context-dep.o] Error 1
    make:     [tree] Error 2

    I already have gcc 4.4 and all tools for Kaldi installed. Any idea what might be the problem? Thanks very much!

     

    • Paul Dixon

      Paul Dixon - 2014-07-15

      Does it work if you use the default OSX compiler(clang)?

      On 15 July 2014 07:40, enzyme156 enzyme156@users.sf.net wrote:

      Hello,

      I am having problems installing Kaldi on my Mac OS X Mavericks. I have
      successfully installed everything required in /tools using tips from this
      thread: https://sourceforge.net/p/kaldi/mailman/message/31618662/. Now I
      am moving onto /src. ./configure and make depend worked fine, but make
      returned this error:

      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
      make[1]: Nothing to be done for all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
      Nothing to be done forall'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
      make[1]: Nothing to be done for `all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
      g++ -msse -msse2 -Wall -I.. -DKALDI_DOUBLEPRECISION=0
      -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1
      -DHAVE_CXXABI_H -rdynamic -DHAVE_CLAPACK
      -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
      context-dep.o context-dep.cc
      g++: error: unrecognized command line option '-rdynamic'
      make[1]:
      * [context-dep.o] Error 1 make: * [tree] Error 2

      Editing kaldi.mk with the version suggested for Mac OS X 10.9 from the
      thread linked above I also got an error:

      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
      make[1]: Nothing to be done for all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
      Nothing to be done forall'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
      make[1]: Nothing to be done for `all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
      c++ -msse -msse2 -std=c++11 -stdlib=libstdc++ -Wall -I.. -fPIC
      -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare
      -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -DHAVE_CLAPACK
      -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
      context-dep.o context-dep.cc
      c++: error: unrecognized command line option '-stdlib=libstdc++'
      make[1]:
      * [context-dep.o] Error 1 make: * [tree] Error 2

      Removing '-stdlib=libstdc++' just to see what happens gives:

      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
      make[1]: Nothing to be done for all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
      Nothing to be done forall'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
      make[1]: Nothing to be done for `all'.
      /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
      c++ -msse -msse2 -Wall -I.. -fPIC -DKALDI_DOUBLEPRECISION=0
      -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1
      -DHAVE_CXXABI_H -DHAVE_CLAPACK
      -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
      context-dep.o context-dep.cc
      In file included from /opt/local/include/gcc48/c++/unordered_map:35:0,
      from ../util/stl-utils.h:32,
      from ../tree/event-map.h:25,
      from ../tree/context-dep.h:22,
      from context-dep.cc:18:
      /opt/local/include/gcc48/c++/bits/c++0x_warning.h:32:2: error: #error This
      file requires compiler and library support for the ISO C++ 2011 standard.
      This support is currently experimental, and must be enabled with the
      -std=c++11 or -std=gnu++11 compiler options.

      error This file requires compiler and library support for the \ ^

      In file included from ../tree/event-map.h:25:0,
      from ../tree/context-dep.h:22,
      from context-dep.cc:18:
      ../util/stl-utils.h:37:12: error: 'std::unordered_map' has not been
      declared
      using std::unordered_map;
      ^
      ../util/stl-utils.h:38:12: error: 'std::unordered_set' has not been
      declared
      using std::unordered_set;
      ^
      In file included from
      /System/Library/Frameworks/vecLib.framework/Headers/vecLib.h:31:0,
      from
      /System/Library/Frameworks/Accelerate.framework/Headers/Accelerate.h:20,
      from ../matrix/kaldi-blas.h:56,
      from ../matrix/matrix-common.h:25,
      from ../matrix/kaldi-vector.h:24,
      from ../matrix/matrix-lib.h:24,
      from ../tree/context-dep.h:23,
      from context-dep.cc:18:
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt16 vU16HalfMultiply(vUInt16, vUInt16)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84:
      note: use -flax-vector-conversions to permit conversions between vectors
      with differing element types or numbers of subparts
      vUInt16 vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA,

      vbasicops_vB); } ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84:
      error: cannot convert 'vUInt16 {aka       vector(8) short unsigned int}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_mullo_epi16(      m128i,

      m128i)' /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:
      In function 'vSInt16 vS16HalfMultiply(vSInt16, vSInt16)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:377:84:
      error: cannot convert 'vSInt16 {aka       vector(8) short int}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_mullo_epi16(      m128i,

      m128i)' vSInt16 vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA, vbasicops_vB);
      }
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vU32FullMulEven(vUInt32, vUInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:432:55:
      error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
      {aka vector(2) long long int}' for argument '1' to 'm128i
      _mm_srli_epi64(

      m128i, int)' vbasicops_vA = _mm_srli_epi64(

      vbasicops_vA, 32); ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:433:55:
      error: cannot convert 'vUInt32 {aka       vector(4) unsigned int}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_srli_epi64(      m128i, int)'
      vbasicops_vB = _mm_srli_epi64(vbasicops_vB, 32);
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:434:56:
      error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
      {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(m128i,
      m128i)'
      return _mm_mul_epu32(vbasicops_vA, vbasicops_vB);
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vU32FullMulOdd(vUInt32, vUInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:459:82:
      error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
      {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(m128i,
      m128i)'
      vUInt32 vbasicops_vB) { return _mm_mul_epu32(vbasicops_vA,

      vbasicops_vB); } ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vU64Sub(vUInt32, vUInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:628:82:
      error: cannot convert 'vUInt32 {aka       vector(4) unsigned int}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_sub_epi64(      m128i,

      m128i)' vUInt32 vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA, vbasicops_vB);
      }
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vSInt32 vS64Sub(vSInt32, vSInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:698:82:
      error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka vector(2)
      long long int}' for argument '1' to '      m128i _mm_sub_epi64(m128i, m128i)'
      vSInt32 vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA,

      vbasicops_vB); } ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vU64Add(vUInt32, vUInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:765:82:
      error: cannot convert 'vUInt32 {aka       vector(4) unsigned int}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_add_epi64(      m128i,

      m128i)' vUInt32 vbasicops_vB) { return _mm_add_epi64(vbasicops_vA, vbasicops_vB);
      }
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vSInt32 vS64Add(vSInt32, vSInt32)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:830:82:
      error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka vector(2)
      long long int}' for argument '1' to '      m128i _mm_add_epi64(m128i, m128i)'
      vSInt32 vbasicops_vB) { return _mm_add_epi64(vbasicops_vA,

      vbasicops_vB); } ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vLL64Shift(vUInt32, vUInt8)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:930:91:
      error: cannot convert 'vUInt8 {aka       vector(16) unsigned char}' to 'm128i
      {aka       vector(2) long long int}' for argument '1' to 'm128i
      _mm_and_si128(      m128i,
      m128i)' _mm_and_si128(vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F
      )));
      ^
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
      function 'vUInt32 vLR64Shift(vUInt32, vUInt8)':
      /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:968:91:
      error: cannot convert 'vUInt8 {aka vector(16) unsigned char}' to 'm128i
      {aka vector(2) long long int}' for argument '1' to 'm128i _mm_and_si128(m128i,
      m128i)'
      _mm_and_si128(__vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F )));
      ^
      In file included from ../base/kaldi-common.h:36:0,
      from ../itf/context-dep-itf.h:21,
      from ../tree/context-dep.h:21,
      from context-dep.cc:18:
      ../base/kaldi-math.h: At global scope:
      ../base/kaldi-math.h:84:17: warning: 'kaldi::kBaseLogZero' defined but not
      used [-Wunused-variable]
      const BaseFloat kBaseLogZero = -std::numeric_limits<BaseFloat>::infinity();
      ^
      ../base/kaldi-math.h:87:17: warning: 'kaldi::kBaseFloatMax' defined but
      not used [-Wunused-variable]
      const BaseFloat kBaseFloatMax = std::numeric_limits<BaseFloat>::max();
      ^
      make[1]:
      * [context-dep.o] Error 1 make: * [tree] Error 2

      I already have gcc 4.4 and all tools for Kaldi installed. Any idea what
      might be the problem? Thanks very much!

      Installing Kaldi on Mac OS X 10.9 (Mavericks)
      https://sourceforge.net/p/kaldi/discussion/1355348/thread/872591a4/?limit=50#e7e8

      Sent from sourceforge.net because you indicated interest in
      https://sourceforge.net/p/kaldi/discussion/1355348/

      To unsubscribe from further messages, please visit
      https://sourceforge.net/auth/subscriptions/

       

      • Daniel Povey

        Daniel Povey - 2014-07-15

        I am a little surprised your compilation options end up with -rdynamic in
        them, because the configure script checks the version of OS X, and selects
        the appropriate one; for Mavericks (OS X 10.9), it should select makefiles/
        darwin_10_9.mk which does not have the -rdynamic option.
        See if you can figure out what "configure" is doing.
        It is supposed to work out of the box.
        Dan

        On Mon, Jul 14, 2014 at 10:43 PM, Paul Dixon edobashira@users.sf.net
        wrote:

        Does it work if you use the default OSX compiler(clang)?

        On 15 July 2014 07:40, enzyme156 enzyme156@users.sf.net wrote:

        Hello,

        I am having problems installing Kaldi on my Mac OS X Mavericks. I have
        successfully installed everything required in /tools using tips from this
        thread: https://sourceforge.net/p/kaldi/mailman/message/31618662/. Now I
        am moving onto /src. ./configure and make depend worked fine, but make
        returned this error:

        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
        make[1]: Nothing to be done for all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
        Nothing to be done forall'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
        make[1]: Nothing to be done for `all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
        g++ -msse -msse2 -Wall -I.. -DKALDI_DOUBLEPRECISION=0
        -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1
        -DHAVE_CXXABI_H -rdynamic -DHAVE_CLAPACK
        -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
        context-dep.o context-dep.cc
        g++: error: unrecognized command line option '-rdynamic'
        make[1]:
        * [context-dep.o] Error 1 make: * [tree] Error 2

        Editing kaldi.mk with the version suggested for Mac OS X 10.9 from the
        thread linked above I also got an error:

        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
        make[1]: Nothing to be done for all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
        Nothing to be done forall'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
        make[1]: Nothing to be done for `all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
        c++ -msse -msse2 -std=c++11 -stdlib=libstdc++ -Wall -I.. -fPIC
        -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare
        -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -DHAVE_CLAPACK
        -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
        context-dep.o context-dep.cc
        c++: error: unrecognized command line option '-stdlib=libstdc++'
        make[1]:
        * [context-dep.o] Error 1 make: * [tree] Error 2

        Removing '-stdlib=libstdc++' just to see what happens gives:

        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
        make[1]: Nothing to be done for all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix make[1]:
        Nothing to be done forall'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C util
        make[1]: Nothing to be done for `all'.
        /Applications/Xcode.app/Contents/Developer/usr/bin/make -C tree
        c++ -msse -msse2 -Wall -I.. -fPIC -DKALDI_DOUBLEPRECISION=0
        -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1
        -DHAVE_CXXABI_H -DHAVE_CLAPACK
        -I/Users/enzyme156/desktop/kaldi-stable/tools/openfst/include -g -c -o
        context-dep.o context-dep.cc
        In file included from /opt/local/include/gcc48/c++/unordered_map:35:0,
        from ../util/stl-utils.h:32,
        from ../tree/event-map.h:25,
        from ../tree/context-dep.h:22,
        from context-dep.cc:18:
        /opt/local/include/gcc48/c++/bits/c++0x_warning.h:32:2: error: #error This
        file requires compiler and library support for the ISO C++ 2011 standard.
        This support is currently experimental, and must be enabled with the
        -std=c++11 or -std=gnu++11 compiler options.
        error This file requires compiler and library support for the \ ^

        In file included from ../tree/event-map.h:25:0,
        from ../tree/context-dep.h:22,
        from context-dep.cc:18:
        ../util/stl-utils.h:37:12: error: 'std::unordered_map' has not been
        declared
        using std::unordered_map;
        ^
        ../util/stl-utils.h:38:12: error: 'std::unordered_set' has not been
        declared
        using std::unordered_set;
        ^
        In file included from
        /System/Library/Frameworks/vecLib.framework/Headers/vecLib.h:31:0,
        from
        /System/Library/Frameworks/Accelerate.framework/Headers/Accelerate.h:20,
        from ../matrix/kaldi-blas.h:56,
        from ../matrix/matrix-common.h:25,
        from ../matrix/kaldi-vector.h:24,
        from ../matrix/matrix-lib.h:24,
        from ../tree/context-dep.h:23,
        from context-dep.cc:18:
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt16 vU16HalfMultiply(vUInt16, vUInt16)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84:
        note: use -flax-vector-conversions to permit conversions between vectors
        with differing element types or numbers of subparts
        vUInt16 vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA,

        vbasicops_vB); } ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352:84:
        error: cannot convert 'vUInt16 {aka         vector(8) short unsigned int}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_mullo_epi16(m128i,

        m128i)' /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:
        In function 'vSInt16 vS16HalfMultiply(vSInt16, vSInt16)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:377:84:
        error: cannot convert 'vSInt16 {aka         vector(8) short int}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_mullo_epi16(m128i,

        m128i)' vSInt16 vbasicops_vB) { return _mm_mullo_epi16(vbasicops_vA,
        vbasicops_vB);
        }
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vU32FullMulEven(vUInt32, vUInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:432:55:
        error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
        {aka vector(2) long long int}' for argument '1' to 'm128i
        _mm_srli_epi64(

        m128i, int)' vbasicops_vA = _mm_srli_epi64(

        vbasicops_vA, 32); ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:433:55:
        error: cannot convert 'vUInt32 {aka         vector(4) unsigned int}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_srli_epi64(m128i, int)'
        vbasicops_vB = _mm_srli_epi64(vbasicops_vB, 32);
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:434:56:
        error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
        {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(
        m128i, m128i)'
        return _mm_mul_epu32(vbasicops_vA, vbasicops_vB);
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vU32FullMulOdd(vUInt32, vUInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:459:82:
        error: cannot convert 'vUInt32 {aka vector(4) unsigned int}' to 'm128i
        {aka vector(2) long long int}' for argument '1' to 'm128i _mm_mul_epu32(
        m128i, m128i)'
        vUInt32 vbasicops_vB) { return _mm_mul_epu32(vbasicops_vA,

        vbasicops_vB); } ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vU64Sub(vUInt32, vUInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:628:82:
        error: cannot convert 'vUInt32 {aka         vector(4) unsigned int}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_sub_epi64(m128i,

        m128i)' vUInt32 vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA,
        vbasicops_vB);
        }
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vSInt32 vS64Sub(vSInt32, vSInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:698:82:
        error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka
        vector(2) long long int}' for argument '1' to 'm128i _mm_sub_epi64(m128i,
        m128i)'
        vSInt32 vbasicops_vB) { return _mm_sub_epi64(vbasicops_vA,

        vbasicops_vB); } ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vU64Add(vUInt32, vUInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:765:82:
        error: cannot convert 'vUInt32 {aka         vector(4) unsigned int}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_add_epi64(m128i,

        m128i)' vUInt32 vbasicops_vB) { return _mm_add_epi64(vbasicops_vA,
        vbasicops_vB);
        }
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vSInt32 vS64Add(vSInt32, vSInt32)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:830:82:
        error: cannot convert 'vSInt32 {aka vector(4) int}' to 'm128i {aka
        vector(2) long long int}' for argument '1' to 'm128i _mm_add_epi64(m128i,
        m128i)'
        vSInt32 vbasicops_vB) { return _mm_add_epi64(vbasicops_vA,

        vbasicops_vB); } ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vLL64Shift(vUInt32, vUInt8)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:930:91:
        error: cannot convert 'vUInt8 {aka         vector(16) unsigned char}' to '
        m128i {aka vector(2) long long int}' for argument '1' to '
        m128i _mm_and_si128(m128i,
        m128i)' _mm_and_si128(vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F
        )));
        ^
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In
        function 'vUInt32 vLR64Shift(vUInt32, vUInt8)':
        /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:968:91:
        error: cannot convert 'vUInt8 {aka vector(16) unsigned char}' to 'm128i
        {aka vector(2) long long int}' for argument '1' to 'm128i _mm_and_si128(
        m128i, m128i)'
        _mm_and_si128(__vbasicops_vShiftFactor, _mm_cvtsi32_si128( 0x3F )));
        ^
        In file included from ../base/kaldi-common.h:36:0,
        from ../itf/context-dep-itf.h:21,
        from ../tree/context-dep.h:21,
        from context-dep.cc:18:
        ../base/kaldi-math.h: At global scope:
        ../base/kaldi-math.h:84:17: warning: 'kaldi::kBaseLogZero' defined but not
        used [-Wunused-variable]
        const BaseFloat kBaseLogZero = -std::numeric_limits<BaseFloat>::infinity();
        ^
        ../base/kaldi-math.h:87:17: warning: 'kaldi::kBaseFloatMax' defined but
        not used [-Wunused-variable]
        const BaseFloat kBaseFloatMax = std::numeric_limits<BaseFloat>::max();
        ^
        make[1]:
        * [context-dep.o] Error 1 make: * [tree] Error 2

        I already have gcc 4.4 and all tools for Kaldi installed. Any idea what
        might be the problem? Thanks very much!

        Installing Kaldi on Mac OS X 10.9 (Mavericks)

        https://sourceforge.net/p/kaldi/discussion/1355348/thread/872591a4/?limit=50#e7e8

        Sent from sourceforge.net because you indicated interest in
        https://sourceforge.net/p/kaldi/discussion/1355348/

        To unsubscribe from further messages, please visit
        https://sourceforge.net/auth/subscriptions/

        Installing Kaldi on Mac OS X 10.9 (Mavericks)
        http://sourceforge.net/p/kaldi/discussion/1355348/thread/872591a4/?limit=50#e7e8/b956

        Sent from sourceforge.net because you indicated interest in
        https://sourceforge.net/p/kaldi/discussion/1355348/

        To unsubscribe from further messages, please visit
        https://sourceforge.net/auth/subscriptions/

         

        • enzyme156

          enzyme156 - 2014-07-16

          The version of Kaldi I am trying to install is kaldi-stable, and it did not come with a .mk file for Mac OS X 10.9. I had to manually create one...

           

          • Jan "yenda" Trmal

            Jan "yenda" Trmal - 2014-07-16

            I think you should not use stable, unless you really know you want/need to
            use it.
            That actually explains your troubles. Use the "current" version - its
            reliable for almost any usage.
            Y.
            On Jul 15, 2014 10:56 PM, "enzyme156" enzyme156@users.sf.net wrote:

            The version of Kaldi I am trying to install is kaldi-stable, and it did
            not come with a .mk file for Mac OS X 10.9. I had to manually create one...

            Installing Kaldi on Mac OS X 10.9 (Mavericks)
            https://sourceforge.net/p/kaldi/discussion/1355348/thread/872591a4/?limit=25#e7e8/b956/91ea/8f1d

            Sent from sourceforge.net because you indicated interest in
            https://sourceforge.net/p/kaldi/discussion/1355348/

            To unsubscribe from further messages, please visit
            https://sourceforge.net/auth/subscriptions/

             

            • enzyme156

              enzyme156 - 2014-07-16

              Thank you for your reply. I downloaded kaldi-trunk and am trying to install it, but running make in the /tools folder gave this error:

              aclocal: macro _LT_DECL_SED' required but not defined aclocal: macro_LT_FUNC_STRIPNAME_CNF' required but not defined
              aclocal failed
              make: *** [irstlm/Makefile] Error 1

              Do you know what the problem might be?

               

              • Jan "yenda" Trmal

                Jan "yenda" Trmal - 2014-07-16

                That looks like as if you have some problems with automake tools (old
                version probably). First, figure out which tool cannot be build and then
                try googling or update the automake tools if nothing else helps.
                Y.
                On Jul 15, 2014 11:35 PM, "enzyme156" enzyme156@users.sf.net wrote:

                Thank you for your reply. I downloaded kaldi-trunk and am trying to
                install it, but running make in the /tools folder gave this error:

                aclocal: macro _LT_DECL_SED' required but not defined aclocal: macro_LT_FUNC_STRIPNAME_CNF'
                required but not defined
                aclocal failed
                make: *** [irstlm/Makefile] Error 1

                Do you know what the problem might be?

                Installing Kaldi on Mac OS X 10.9 (Mavericks)
                https://sourceforge.net/p/kaldi/discussion/1355348/thread/872591a4/?limit=25#e7e8/b956/91ea/8f1d/6fe3/2c01

                Sent from sourceforge.net because you indicated interest in
                https://sourceforge.net/p/kaldi/discussion/1355348/

                To unsubscribe from further messages, please visit
                https://sourceforge.net/auth/subscriptions/

                 

      • enzyme156

        enzyme156 - 2014-07-16

        Silly question but... how do I do that? Thank you!

         

  • enzyme156

    enzyme156 - 2014-07-16

    I used kaldi-trunk instead and installation was successful. Thank you all for your help! :)

     

  • enzyme156

    enzyme156 - 2014-07-24

    Hello,

    Sorry to bother you again, but after a few problems with my computer I had to reinstall Kaldi, and running make in /src gave this:

    The version of configure script matches kaldi.mk version. Good.
    [ "1.3.4" == '1.3.2' ] || [ "1.3.4" == '1.3.3' ] || [ "1.3.4" == '1.3.4' ] || { echo "You now need openfst-1.3.2 or later. cd ../tools; svn update; ./install.sh; cd ../src; make depend; make"; exit 1; };
    test -d || mkdir
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C base
    make[1]: Nothing to be done for `all'.
    /Applications/Xcode.app/Contents/Developer/usr/bin/make -C matrix
    g++ -msse -msse2 -Wall -I.. -fPIC -DKALDI_DOUBLEPRECISION=0 -DHAVE_POSIX_MEMALIGN -Wno-sign-compare -Winit-self -DHAVE_EXECINFO_H=1 -DHAVE_CXXABI_H -DHAVE_CLAPACK -I/Users/anh/Desktop/fpt/kaldi-trunk/tools/openfst/include -g -c -o kaldi-matrix.o kaldi-matrix.cc
    In file included from /System/Library/Frameworks/vecLib.framework/Headers/vecLib.h:31,
    from /System/Library/Frameworks/Accelerate.framework/Headers/Accelerate.h:20,
    from ../matrix/kaldi-blas.h:58,
    from ../matrix/matrix-common.h:27,
    from ../matrix/kaldi-matrix.h:25,
    from kaldi-matrix.cc:23:
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'short unsigned int vector vU16HalfMultiply(short unsigned int vector, short unsigned int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352: note: use -flax-vector-conversions to permit conversions between vectors with differing element types or numbers of subparts
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:352: error: cannot convert 'short unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_mullo_epi16(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'short int vector vS16HalfMultiply(short int vector, short int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:377: error: cannot convert 'short int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_mullo_epi16(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vU32FullMulEven(unsigned int vector, unsigned int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:432: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_srli_epi64(long long int vector, int)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:433: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_srli_epi64(long long int vector, int)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:434: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_mul_epu32(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vU32FullMulOdd(unsigned int vector, unsigned int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:459: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_mul_epu32(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vU64Sub(unsigned int vector, unsigned int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:628: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_sub_epi64(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'int vector vS64Sub(int vector, int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:698: error: cannot convert 'int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_sub_epi64(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vU64Add(unsigned int vector, unsigned int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:765: error: cannot convert 'unsigned int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_add_epi64(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'int vector vS64Add(int vector, int vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:830: error: cannot convert 'int vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_add_epi64(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vLL64Shift(unsigned int vector, unsigned char vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:930: error: cannot convert 'unsigned char vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_and_si128(long long int vector, long long int vector)'
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h: In function 'unsigned int vector vLR64Shift(unsigned int vector, unsigned char vector)':
    /System/Library/Frameworks/vecLib.framework/Headers/vBasicOps.h:968: error: cannot convert 'unsigned char vector' to 'long long int vector' for argument '1' to 'long long int vector _mm_and_si128(long long int vector, long long int vector)'
    ../base/kaldi-math.h: At global scope:
    ../base/kaldi-math.h:86: warning: 'kaldi::kLogZeroBaseFloat' defined but not used
    make[1]: [kaldi-matrix.o] Error 1
    make:     [matrix] Error 2

    make in /tools and ./configure and make depend in /src were successful, so what could the problem have been? My version of openfst is openfst-1.3.4 if that helps.