Allow for using .txt files in aligning

Resolves #94

aligner/corpus.py

@@ -20,7 +20,7 @@
 
 def find_lab(filename, files):
     '''
-    Finds a .lab file that corresponds to a wav file
+    Finds a .lab file or .txt file that corresponds to a wav file.  The .lab extension is given priority.
 
     Parameters
     ----------
@@ -32,13 +32,17 @@ def find_lab(filename, files):
     Returns
     -------
     str or None
-        If a corresponding .lab file is found, returns it, otherwise returns None
+        If a corresponding .lab or .txt file is found, returns it, otherwise returns None
     '''
     name, ext = os.path.splitext(filename)
     for f in files:
         fn, fext = os.path.splitext(f)
         if fn == name and fext.lower() == '.lab':
             return f
+    for f in files: # Use .txt if no .lab file available
+        fn, fext = os.path.splitext(f)
+        if fn == name and fext.lower() == '.txt':
+            return f
     return None
 
 

docs/source/data_format.rst

@@ -20,6 +20,8 @@ text file with the .lab extension).
    you can generate .lab files from it using the relabel function of relabel_clean.py.
    The relabel_clean.py script is currently in the prosodylab.alignertools repository on GitHub.
 
+If no ``.lab`` file is found, then the aligner will look for any matching ``.txt`` files and use those.
+
 In terms of directory structure, the default configuration assumes that
 files are separated into subdirectories based on their speaker (with one
 speaker per file).
@@ -66,7 +68,8 @@ each speaker.
         :align: center
         :alt: Image cannot be displayed in your browser
 
-.. note ::
+.. note::
+
    Intervals in the TextGrid less than 100 milliseconds will not be aligned.
 
 Transcription normalization and dictionary lookup

tests/conftest.py

@@ -70,6 +70,20 @@ def basic_corpus_dir(corpus_root_dir, wav_dir, lab_dir):
     return path
 
 
+@pytest.fixture(scope='session')
+def basic_corpus_txt_dir(corpus_root_dir, wav_dir, lab_dir):
+    path = os.path.join(corpus_root_dir, 'basic_txt')
+    os.makedirs(path, exist_ok=True)
+    names = [('michael', ['acoustic_corpus']), ('sickmichael', ['cold_corpus', 'cold_corpus3'])]
+    for s, files in names:
+        s_dir = os.path.join(path, s)
+        os.makedirs(s_dir, exist_ok=True)
+        for name in files:
+            shutil.copyfile(os.path.join(wav_dir, name + '.wav'), os.path.join(s_dir, name + '.wav'))
+            shutil.copyfile(os.path.join(lab_dir, name + '.lab'), os.path.join(s_dir, name + '.txt'))
+    return path
+
+
 @pytest.fixture(scope='session')
 def extra_corpus_dir(corpus_root_dir, wav_dir, lab_dir):
     path = os.path.join(corpus_root_dir, 'extra')

tests/test_corpus.py

@@ -12,11 +12,23 @@ def test_basic(basic_dict_path, basic_corpus_dir, generated_dir):
     dictionary = Dictionary(basic_dict_path, os.path.join(generated_dir, 'basic'))
     dictionary.write()
     output_directory = os.path.join(generated_dir, 'basic')
-    d = Corpus(basic_corpus_dir, output_directory)
-    d.initialize_corpus(dictionary)
+    c = Corpus(basic_corpus_dir, output_directory)
+    c.initialize_corpus(dictionary)
     fc = FeatureConfig()
-    fc.generate_features(d)
-    assert d.get_feat_dim(fc) == 39
+    fc.generate_features(c)
+    assert c.get_feat_dim(fc) == 39
+
+
+def test_basic_txt(basic_corpus_txt_dir, basic_dict_path, generated_dir):
+    dictionary = Dictionary(basic_dict_path, os.path.join(generated_dir, 'basic'))
+    dictionary.write()
+    output_directory = os.path.join(generated_dir, 'basic')
+    c = Corpus(basic_corpus_txt_dir, output_directory)
+    assert len(c.no_transcription_files) == 0
+    c.initialize_corpus(dictionary)
+    fc = FeatureConfig()
+    fc.generate_features(c)
+    assert c.get_feat_dim(fc) == 39
 
 
 def test_extra(sick_dict, extra_corpus_dir, generated_dir):