summaryrefslogtreecommitdiff
path: root/pkgs/development/python-modules/pyannote-audio/default.nix
blob: ea4a25888d37a6d17b2f0186f54111022dd0e25e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
{
  lib,
  stdenv,
  buildPythonPackage,
  fetchFromGitHub,

  # build-system
  hatchling,
  hatch-vcs,

  # dependencies
  asteroid-filterbanks,
  einops,
  huggingface-hub,
  lightning,
  matplotlib,
  opentelemetry-api,
  opentelemetry-exporter-otlp,
  opentelemetry-sdk,
  pyannote-core,
  pyannote-database,
  pyannote-metrics,
  pyannote-pipeline,
  pyannoteai-sdk,
  pytorch-metric-learning,
  safetensors,
  soundfile,
  speechbrain,
  tensorboardx,
  torch,
  torch-audiomentations,
  torchaudio,
  torchcodec,
  torchmetrics,

  # optional-dependencies
  hydra-core,
  typer,

  # tests
  papermill,
  pytestCheckHook,
}:

buildPythonPackage rec {
  pname = "pyannote-audio";
  version = "4.0.1";
  pyproject = true;

  src = fetchFromGitHub {
    owner = "pyannote";
    repo = "pyannote-audio";
    tag = version;
    fetchSubmodules = true;
    hash = "sha256-hYrwpph+Powt+AuQjKo0kkBW+5jJGfzGTILzL9j22YI=";
  };

  build-system = [
    hatchling
    hatch-vcs
  ];

  pythonRelaxDeps = [
    "torchaudio"
  ];
  dependencies = [
    asteroid-filterbanks
    einops
    huggingface-hub
    lightning
    matplotlib
    opentelemetry-api
    opentelemetry-exporter-otlp
    opentelemetry-sdk
    pyannote-core
    pyannote-database
    pyannote-metrics
    pyannote-pipeline
    pyannoteai-sdk
    pytorch-metric-learning
    safetensors
    soundfile
    speechbrain
    tensorboardx
    torch
    torch-audiomentations
    torchaudio
    torchcodec
    torchmetrics
  ];

  optional-dependencies = {
    cli = [
      hydra-core
      typer
    ];
  };

  pythonImportsCheck = [ "pyannote.audio" ];

  nativeCheckInputs = [
    papermill
    pytestCheckHook
  ];

  preCheck = ''
    $out/bin/pyannote-audio --help
  '';

  disabledTests = [
    # Require internet access
    "test_hf_download_inference"
    "test_hf_download_model"
    "test_import_speechbrain_encoder_classifier"
    "test_skip_aggregation"
    "test_unknown_specifications_error_raised_on_non_setup_model_task"

    # AttributeError: module 'torchaudio' has no attribute 'info'
    # Removed in torchaudio v2.9.0
    # See https://github.com/pytorch/audio/issues/3902 for context
    "test_audio_resample"
  ];

  disabledTestPaths = lib.optionals stdenv.hostPlatform.isDarwin [
    # Crashes the interpreter
    # - On aarch64-darwin: Trace/BPT trap: 5
    # - On x86_64-darwin: Fatal Python error: Illegal instruction
    "tests/inference_test.py"
    "tests/test_train.py"
  ];

  meta = {
    description = "Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding";
    homepage = "https://github.com/pyannote/pyannote-audio";
    changelog = "https://github.com/pyannote/pyannote-audio/blob/${src.tag}/CHANGELOG.md";
    license = lib.licenses.mit;
    maintainers = with lib.maintainers; [
      GaetanLepage
    ];
  };
}