From 2c552f9c10aad76d7358d5c89f60bd876a780a22 Mon Sep 17 00:00:00 2001
From: chrisxu2016 <823254351@qq.com>
Date: Tue, 20 Jun 2017 18:19:43 +0800
Subject: [PATCH] add augmentor class

---
 .../data_utils/augmentor/augmentation.py      |  2 +-
 deep_speech_2/tests/test_augmentor.py         | 60 +++++++++++++++++++
 2 files changed, 61 insertions(+), 1 deletion(-)
 create mode 100755 deep_speech_2/tests/test_augmentor.py

diff --git a/deep_speech_2/data_utils/augmentor/augmentation.py b/deep_speech_2/data_utils/augmentor/augmentation.py
index bfe7075e..08788008 100755
--- a/deep_speech_2/data_utils/augmentor/augmentation.py
+++ b/deep_speech_2/data_utils/augmentor/augmentation.py
@@ -83,7 +83,7 @@ class AugmentationPipeline(object):
             return SpeedPerturbAugmentor(self._rng, **params)
         if augmentor_type == "resample":
             return ResampleAugmentor(self._rng, **params)
-        if augmentor_type == "baysian_normal":
+        if augmentor_type == "bayesian_normal":
             return OnlineBayesianNormalizationAugmentor(self._rng, **params)
         else:
             raise ValueError("Unknown augmentor type [%s]." % augmentor_type)
diff --git a/deep_speech_2/tests/test_augmentor.py b/deep_speech_2/tests/test_augmentor.py
new file mode 100755
index 00000000..76fd321a
--- /dev/null
+++ b/deep_speech_2/tests/test_augmentor.py
@@ -0,0 +1,60 @@
+"""Test augmentor class."""
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import unittest
+from data_utils import audio
+from data_utils.augmentor.augmentation import AugmentationPipeline
+import random
+import numpy as np
+
+random_seed=0
+#audio instance
+audio_data=[3.05175781e-05, -8.54492188e-04, -1.09863281e-03, -9.46044922e-04,\
+            -1.31225586e-03, -1.09863281e-03, -1.73950195e-03, -2.10571289e-03,\
+            -2.04467773e-03, -1.46484375e-03, -1.43432617e-03, -9.46044922e-04,\
+            -1.95312500e-03, -1.86157227e-03, -2.10571289e-03, -2.31933594e-03,\
+            -2.01416016e-03, -2.62451172e-03, -2.07519531e-03, -2.38037109e-03]
+audio_data = np.array(audio_data)
+samplerate = 10
+
+class TestAugmentor(unittest.TestCase):
+    def test_volume(self):
+        augmentation_config='[{"type": "volume","params": {"min_gain_dBFS": -15, "max_gain_dBFS": 15},"prob": 1.0}]'
+        augmentation_pipeline = AugmentationPipeline(augmentation_config=augmentation_config,
+                                                     random_seed=random_seed)
+        audio_segment = audio.AudioSegment(audio_data, samplerate)
+        augmentation_pipeline.transform_audio(audio_segment)
+        original_audio = audio.AudioSegment(audio_data, samplerate)
+        self.assertFalse(np.any(audio_segment.samples == original_audio.samples))
+
+    def test_speed(self):
+        augmentation_config='[{"type": "speed","params": {"min_speed_rate": 1.2,"max_speed_rate": 1.4},"prob": 1.0}]'
+        augmentation_pipeline = AugmentationPipeline(augmentation_config=augmentation_config,
+                                                     random_seed=random_seed)
+        audio_segment = audio.AudioSegment(audio_data, samplerate)
+        augmentation_pipeline.transform_audio(audio_segment)
+        original_audio = audio.AudioSegment(audio_data, samplerate)
+        self.assertFalse(np.any(audio_segment.samples == original_audio.samples))
+
+    def test_resample(self):
+        augmentation_config='[{"type": "resample","params": {"new_sample_rate":5},"prob": 1.0}]'
+        augmentation_pipeline = AugmentationPipeline(augmentation_config=augmentation_config,
+                                                     random_seed=random_seed)
+        audio_segment = audio.AudioSegment(audio_data, samplerate)
+        augmentation_pipeline.transform_audio(audio_segment)
+        self.assertTrue(audio_segment.sample_rate == 5)
+
+    def test_bayesial(self):
+        augmentation_config='[{"type": "bayesian_normal","params": {"target_db": -20, "prior_db": -4, "prior_samples": -8, "startup_delay": 0.0},"prob": 1.0}]'
+        augmentation_pipeline = AugmentationPipeline(augmentation_config=augmentation_config,
+                                                     random_seed=random_seed)
+        audio_segment = audio.AudioSegment(audio_data, samplerate)
+        augmentation_pipeline.transform_audio(audio_segment)
+        original_audio = audio.AudioSegment(audio_data, samplerate)
+        self.assertFalse(np.any(audio_segment.samples == original_audio.samples))
+
+if __name__ == '__main__':
+    unittest.main()
+
-- 
GitLab