Add stubs for qwen_tts and torchcodec modules in audio test configuration

Author: sergioburdisso

tests/conftest.py

@@ -0,0 +1,115 @@
+import sys
+import types
+import importlib.machinery
+from types import SimpleNamespace
+
+import numpy as np
+
+
+def _install_qwen_tts_stub() -> None:
+    try:
+        __import__("qwen_tts")
+        return
+    except ImportError:
+        pass
+
+    class _FakeQwen3TTSModel:
+        def __init__(self, *args, **kwargs):
+            self.args = args
+            self.kwargs = kwargs
+
+        @classmethod
+        def from_pretrained(cls, *args, **kwargs):
+            return cls(*args, **kwargs)
+
+        def generate_custom_voice(self, text, speaker=None, **kwargs):
+            return [np.zeros(24_000, dtype=np.float32)], 24_000
+
+        def generate_voice_clone(self, text, **kwargs):
+            return [np.zeros(24_000, dtype=np.float32)], 24_000
+
+        def generate_voice_design(self, text, language=None, instruct=None, **kwargs):
+            return [np.zeros(24_000, dtype=np.float32)], 24_000
+
+        def create_voice_clone_prompt(self, ref_audio=None, ref_text=None, **kwargs):
+            return {
+                "ref_audio": ref_audio,
+                "ref_text": ref_text,
+            }
+
+    qwen_tts_module = types.ModuleType("qwen_tts")
+    qwen_tts_module.Qwen3TTSModel = _FakeQwen3TTSModel
+    qwen_tts_module.__spec__ = importlib.machinery.ModuleSpec("qwen_tts", loader=None)
+    sys.modules["qwen_tts"] = qwen_tts_module
+
+
+_install_qwen_tts_stub()
+
+
+def _install_torchcodec_stub() -> None:
+    try:
+        __import__("torchcodec")
+        return
+    except ImportError:
+        pass
+
+    class _FakeTensor:
+        def __init__(self, array):
+            self._array = array
+
+        def cpu(self):
+            return self
+
+        def numpy(self):
+            return self._array
+
+    class _FakeAudioSamples:
+        def __init__(self, data=None, sample_rate: int = 16_000):
+            _arr = np.zeros((1, sample_rate), dtype=np.float32) if data is None else data
+            self.data = _FakeTensor(_arr)
+            self.sample_rate = sample_rate
+
+    class _FakeAudioDecoder:
+        def __init__(self, source=None, *args, **kwargs):
+            self.source = source
+            self.args = args
+            self.kwargs = kwargs
+            _path = None
+            if isinstance(source, dict):
+                _path = source.get("path")
+            else:
+                _path = getattr(source, "path", None)
+
+            self.metadata = SimpleNamespace(
+                sample_rate=16_000,
+                path=_path,
+            )
+
+        def __getitem__(self, key: str):
+            if key == "path":
+                return self.metadata.path
+            if key == "sampling_rate":
+                return self.metadata.sample_rate
+            if key == "array":
+                y = self.get_all_samples().data.cpu().numpy()
+                return np.mean(y, axis=tuple(range(y.ndim - 1))) if y.ndim > 1 else y
+            raise KeyError(key)
+
+        def get_all_samples(self):
+            return _FakeAudioSamples()
+
+        def get_samples_played_in_range(self, *_args, **_kwargs):
+            return SimpleNamespace(sample_rate=self.metadata.sample_rate)
+
+    torchcodec_module = types.ModuleType("torchcodec")
+    decoders_module = types.ModuleType("torchcodec.decoders")
+    decoders_module.AudioDecoder = _FakeAudioDecoder
+    torchcodec_module.decoders = decoders_module
+    torchcodec_module.__spec__ = importlib.machinery.ModuleSpec("torchcodec", loader=None)
+    decoders_module.__spec__ = importlib.machinery.ModuleSpec("torchcodec.decoders", loader=None)
+
+    sys.modules["torchcodec"] = torchcodec_module
+    sys.modules["torchcodec.decoders"] = decoders_module
+
+
+_install_torchcodec_stub()

tests/test_audio.py

@@ -6,34 +6,34 @@
 import pandas as pd
 
 # Try to import audio dependencies
-try:
-    import soundfile as sf
-
-    from sdialog.audio.turn import AudioTurn
-    from sdialog.audio.room_generator import BasicRoomGenerator
-    from sdialog.audio.utils import Role, AudioUtils, Furniture, SpeakerSide
-    from sdialog.audio.room import Position3D, Dimensions3D, DirectivityType, Room
-    from sdialog.audio.voice_database import Voice, is_a_audio_file
-    from sdialog.audio.voice_database import BaseVoiceDatabase, LocalVoiceDatabase, VoiceDatabase
-    from sdialog.audio.tts import BaseTTS
-    from sdialog.audio.jsalt import MedicalRoomGenerator, RoomRole
-    from sdialog.audio.acoustics_simulator import AcousticsSimulator, AudioSource
-    from sdialog.audio.dialog import AudioDialog
-    from sdialog.audio.pipeline import AudioPipeline, to_audio
-    from sdialog.audio.dscaper_utils import send_utterances_to_dscaper, generate_dscaper_timeline
-    from sdialog.audio.impulse_response_database import LocalImpulseResponseDatabase, RecordingDevice
-    from sdialog.audio.processing import AudioProcessor
-except ImportError:
-    print("\n" + "=" * 80)
-    print("Audio dependencies are not installed. All audio tests will be skipped.")
-    print("=" * 80 + "\n")
-
-    # Skip the entire module - pytest will not collect any tests from this file
-    pytest.skip(
-        "Audio dependencies not installed. If you are working with audio, install them with: "
-        "pip install sdialog[audio]",
-        allow_module_level=True
-    )
+# try:
+import soundfile as sf
+
+from sdialog.audio.turn import AudioTurn
+from sdialog.audio.room_generator import BasicRoomGenerator
+from sdialog.audio.utils import Role, AudioUtils, Furniture, SpeakerSide
+from sdialog.audio.room import Position3D, Dimensions3D, DirectivityType, Room
+from sdialog.audio.voice_database import Voice, is_a_audio_file
+from sdialog.audio.voice_database import BaseVoiceDatabase, LocalVoiceDatabase, VoiceDatabase
+from sdialog.audio.tts import BaseTTS
+from sdialog.audio.jsalt import MedicalRoomGenerator, RoomRole
+from sdialog.audio.acoustics_simulator import AcousticsSimulator, AudioSource
+from sdialog.audio.dialog import AudioDialog
+from sdialog.audio.pipeline import AudioPipeline, to_audio
+from sdialog.audio.dscaper_utils import send_utterances_to_dscaper, generate_dscaper_timeline
+from sdialog.audio.impulse_response_database import LocalImpulseResponseDatabase, RecordingDevice
+from sdialog.audio.processing import AudioProcessor
+# except ImportError:
+#     print("\n" + "=" * 80)
+#     print("Audio dependencies are not installed. All audio tests will be skipped.")
+#     print("=" * 80 + "\n")
+
+#     # Skip the entire module - pytest will not collect any tests from this file
+#     pytest.skip(
+#         "Audio dependencies not installed. If you are working with audio, install them with: "
+#         "pip install sdialog[audio]",
+#         allow_module_level=True
+#     )
 
 from sdialog import Turn, Dialog
 from unittest.mock import MagicMock, patch