python3Packages.torchaudio: 2.10.0 -> 2.11.0

  buildPythonPackage,

  fetchFromGitHub,

  # nativeBuildInputs

  cmake,

  pkg-config,

  ninja,

  # buildInputs

  pybind11,

  sox,

  torch,

  llvmPackages,

  # build-system

  setuptools,

  # dependencies

  torch,

  torchcodec,

  # tests

  inflect,

  parameterized,

  pytestCheckHook,

  pytorch-lightning,

  scipy,

  sentencepiece,

  unidecode,

  # passthru

  torchaudio,

  cudaSupport ? torch.cudaSupport,

  cudaPackages,

  rocmSupport ? torch.rocmSupport,

  rocmPackages,

}:

buildPythonPackage.override { stdenv = torch.stdenv; } (finalAttrs: {

buildPythonPackage.override { inherit (torch) stdenv; } (finalAttrs: {

  pname = "torchaudio";

  version = "2.10.0";

  version = "2.11.0";

  pyproject = true;

  src = fetchFromGitHub {

    owner = "pytorch";

    repo = "audio";

    tag = "v${finalAttrs.version}";

    hash = "sha256-b1sjHVFXdNFDbdtXWSM2KisSRE/8IbzJI4rvzYQ4UMg=";

    hash = "sha256-TncROn9wfn5HOaIvupS2/KD9JCgwfHyfnbZRc+SiqJ0=";

  };

  patches = [

    ./0001-setup.py-propagate-cmakeFlags.patch

  ];

  postPatch = lib.optionalString rocmSupport ''

    # There is no .info/version-dev, only .info/version

    substituteInPlace cmake/LoadHIP.cmake \

      --replace-fail "/.info/version-dev" "/.info/version"

  '';

  env = {

    # CTC seems to be missing no matter what. No obvious flag to force its compilation.

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_CTC_DECODER = 1;

    # CUDA

    USE_CUDA = cudaSupport;

    TORCH_CUDA_ARCH_LIST = "${lib.concatStringsSep ";" torch.cudaCapabilities}";

    BUILD_SOX = 0;

    BUILD_KALDI = 0;

    BUILD_RNNT = 0;

    BUILD_CTC_DECODER = 0;

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_CUCTC_DECODER = 1;

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_CUDA = 1;

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_MULTIGPU_CUDA = 1;

    # At of 2026-03-27, the default `cudaPackages` version is 12.9.1

    # According to Nvidia, it should support GCC versions up to 14.x:

    # -> https://docs.nvidia.com/cuda/archive/12.9.1/cuda-installation-guide-linux/index.html#host-compiler-support-policy

    # However, PyTorch's *strict* upper bound is 14.0:

    # -> https://github.com/pytorch/pytorch/blob/v2.11.0/torch/utils/cpp_extension.py#L75

    # Hence, the build fails with:

    #   RuntimeError: The current installed version of g++ (14.3.0) is greater than the maximum

    #   required version by CUDA 12.9. Please make sure to use an adequate version of g++

    #   (>=6.0.0, <14.0).

    # Hence, we disable the version check to silence the error:

    TORCH_DONT_CHECK_COMPILER_ABI = cudaSupport;

    # ROCM

    USE_ROCM = rocmSupport;

    PYTORCH_ROCM_ARCH = lib.optionalString rocmSupport torch.gpuTargetString;

    ROCM_PATH = lib.optionalString rocmSupport torch.rocmtoolkit_joined;

    # demucs is not packaged in nixpkgs and is archived anyway

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_MOD_demucs = true;

    # fairseq is unmaintained and broken in nixpkgs

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_MOD_fairseq = true;

    # Fails even on python>3.10 with:

    #   RuntimeError: Test is known to fail for Python 3.10, disabling for now

    #   See: https://github.com/pytorch/audio/pull/2224#issuecomment-1048329450

    TORCHAUDIO_TEST_ALLOW_SKIP_IF_ON_PYTHON_310 = true;

  };

  nativeBuildInputs = [

    cmake

    pkg-config

    ninja

  ]

  ++ lib.optionals cudaSupport [ cudaPackages.cuda_nvcc ]

  ++ lib.optionals rocmSupport (

    with rocmPackages;

    [

      clr

      rocblas

      hipblas

    ]

  );

  build-system = [

    setuptools

  ];

  buildInputs = [

    pybind11

    sox

    torch.cxxdev

  nativeBuildInputs = lib.optionals cudaSupport [

    cudaPackages.cuda_nvcc

  ];

  buildInputs =

    lib.optionals cudaSupport [

      cudaPackages.cuda_cudart

    ]

  ++ lib.optionals torch.stdenv.cc.isClang [ llvmPackages.openmp ];

    ++ lib.optionals torch.stdenv.cc.isClang [

      llvmPackages.openmp

    ];

  dependencies = [ torch ];

  dependencies = [

    torch

    torchcodec

  ];

  preConfigure = lib.optionalString rocmSupport ''

    export ROCM_PATH=${torch.rocmtoolkit_joined}

    export PYTORCH_ROCM_ARCH="${torch.gpuTargetString}"

  '';

  pythonImportsCheck = [ "torchaudio" ];

  dontUseCmakeConfigure = true;

  nativeCheckInputs = [

    inflect

    parameterized

    pytestCheckHook

    pytorch-lightning

    scipy

    sentencepiece

    unidecode

  ];

  doCheck = false; # requires sox backend

  disabledTestPaths = [

    # Require internet access

    "test/integration_tests"

  ];

  pythonImportsCheck = [ "torchaudio" ];

  disabledTests = [

    # AssertionError: Tensor-likes are not close!

    # Mismatched elements: 1070 / 32800 (3.3%)

    "test_amplitude_to_DB"

    "test_amplitude_to_DB_power"

    # Very long to run

    "AutogradCPUTest"

  ];

  passthru.gpuCheck = torchaudio.overridePythonAttrs (old: {

    requiredSystemFeatures = [ "cuda" ];

    env = (old.env or { }) // {

      TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_CUCTC_DECODER = 0;

      TORCHAUDIO_TEST_ALLOW_SKIP_IF_NO_CUDA = 0;

    };

  });

  meta = {

    description = "PyTorch audio library";