Tests

scraibe/.pyannotetoken

@@ -1 +1 @@
-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
+

scraibe/test/test_audio.py

@@ -0,0 +1,99 @@
+import pytest
+#from scraibe import Transcriber
+#from unittest.mock import patch, mock_open
+#import unittest
+#import os
+from .audio import AudioProcessor
+import torch
+
+
+
+
+
+test_waveform = torch.tensor([]).to('cuda')
+test_sr = 16000
+SAMPLE_RATE = 16000
+NORMALIZATION_FACTOR = 32768
+
+
+@pytest.fixture
+def probe_audio_processor():
+    """_summary_
+
+    Returns:
+        _type_: _description_
+    """    
+    return AudioProcessor(test_waveform, test_sr)
+
+
+
+
+
+
+def test_AudioProcessor_init(probe_audio_processor):
+    """_summary_
+
+    Args:
+        probe_audio_processor (_type_): _description_
+    """    
+    assert isinstance(probe_audio_processor, AudioProcessor)
+    assert probe_audio_processor.waveform.device == test_waveform.device
+    assert torch.equal(probe_audio_processor.waveform, test_waveform)
+    assert probe_audio_processor.sr == test_sr
+
+
+
+def test_cut():
+    """_summary_
+    """    
+    waveform = torch.Tensor(10, 3)
+    sr = 16000
+    start = 4
+    end = 7
+    assert AudioProcessor(waveform, sr).cut(start, end).size() == int((end - start) * test_sr)
+
+
+
+""" def test_cut(probe_audio_processor):
+    start = 10
+    end = 100
+    test_segment =  probe_audio_processor.cut(start, end) 
+    print(test_segment)
+    erwartetes_segment = int((end - start) * test_sr)
+    print(test_segment.size())
+    assert len(test_segment) == erwartetes_segment
+ """
+
+
+
+
+
+def test_audio_processor_invalid_sr():
+    """_summary_
+    """    
+    with pytest.raises(ValueError):
+        AudioProcessor(test_waveform, [44100,48000])
+
+
+def test_audio_processor_SAMPLE_RATE():
+    """_summary_
+    """    
+    probe_audio_processor = AudioProcessor(test_waveform)
+    assert probe_audio_processor.sr == SAMPLE_RATE       
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

scraibe/test/test_diarisation.py

@@ -0,0 +1,36 @@
+import pytest
+import os
+from unittest import mock
+from scraibe import Diariser
+
+
+
+@pytest.fixture
+def diariser_instance():
+    with mock.patch.object(Diariser, '_get_token', return_value = 'personal Hugging-Face token')
+        return Diariser('pyannote')
+
+
+
+def test_Diariser_init(diariser_instance):
+    """_summary_
+
+    Args:
+        diariser_instance (_type_): _description_
+    """    
+    assert diariser_instance.model == 'pyannote'
+
+
+
+def test_diarisation_function(diariser_instance):
+    """_summary_
+
+    Args:
+        diariser_instance (_type_): _description_
+    """    
+    with mock.patch.object(diariser_instance.model, 'apply', return_value='diarization_result'):
+        diarization_output = diariser_instance.diarization('example_audio_file.wav')
+        assert diarization_output == 'diarization_result'
+
+
+

scraibe/test/test_transcriber.py

@@ -0,0 +1,29 @@
+import pytest
+from unittest.mock import patch
+from scraibe import Transcriber
+
+
+
+
+@pytest.mark.parametrize("audio_file, expected_transcription",[("path_to_test_audiofile", "test_transcription")] )
+@patch("scraibe.Transcriber.load_model")
+
+def test_transcriber(mock_load_model, audio_file, expected_transcription):
+    """_summary_
+
+    Args:
+        mock_load_model (_type_): _description_
+        audio_file (_type_): _description_
+        expected_transcription (_type_): _description_
+    """    
+    mock_model = mock_load_model.return_value
+    mock_model.transcribe.return_value ={"text": expected_transcription} 
+
+    transcriber = Transcriber.load_model(model="medium")
+
+    transcription_result = transcriber.transcribe(audio=audio_file)
+
+    assert transcription_result == expected_transcription
+
+
+