Voice Lab

https://voice-lab.github.io/VoiceLab/
Admin
Voice Lab is an automated voice analysis software that allows users to measure, manipulate, and...
Tokens:5,860
Snippets:45
Trust Score:9.7
Update:1 month ago
Show doc for...
Tokens:
Raw
### Run Voicelab node from command line using prepare_node() and process()

Source: https://voice-lab.github.io/VoiceLab

These functions form the test harness for the ReverseSoundsNode. They set up test files, prepare the node with a sound file, generate reversed validation sounds, and run the node's process() method to compare results. The final test uses pytest parametrization to iterate over all test files except the first broken one.

```python
# Arrange
def get_test_files():
  number_of_test_files = len(os.listdir(AUDIO_DIR)) - 1  # -1 because the first file is a broken sound
  test_files = sorted(os.listdir(AUDIO_DIR))[1:]
  return number_of_test_files, test_files

```

```python
# Arrange
def prepare_node(test_file):
  file_path = os.path.join(AUDIO_DIR, test_file)
  node = ReverseSoundsNode.ReverseSoundsNode()
  node.args['file_path'] = file_path
  # Load the sound and setup the node
  try:
      sound: parselmouth.Sound = parselmouth.Sound(file_path)
      signal = sound.values
      sampling_rate = sound.sampling_frequency
  except:
      signal = None
      sampling_rate = None
  node.voice = (signal, sampling_rate)
  node.args['voice'] = (signal, sampling_rate)
  print((f"{file_path=}..{sampling_rate=}"))
  return node

```

```python
# Arrange
def get_reversed_test_sound(test_file):
  file_path = os.path.join(AUDIO_DIR, test_file)
  validation_sound = parselmouth.Sound(file_path)
  validation_sound.reverse()
  return validation_sound.values

```

```python
# Arrange
def get_number_of_test_files():
  number_of_test_files, _ = get_test_files()
  return number_of_test_files

```

```python
# Arrange
def generate_numpy_arrays(execution_number):
  number_of_test_files, test_files = get_test_files()
  filename = test_files[execution_number]
  node = prepare_node(filename)
  # Run the node
  results = node.process()
  print(results)
  # Validate the results
  validation_sound = get_reversed_test_sound(filename)
  test_sound = results['voice'].values
  return test_sound, validation_sound

# Act
@pytest.mark.parametrize('execution_number', range(len(os.listdir(AUDIO_DIR)) - 1))
def test_ReverseSoundsNode(execution_number):
  test_sound, validation_sound = generate_numpy_arrays(execution_number)
  assert np.array_equal(test_sound, validation_sound)

```
Tokens:
Raw