fix conflict

Author: azraelkuan

tensorflow_tts/bin/preprocess.py

@@ -30,7 +30,9 @@
 from sklearn.preprocessing import StandardScaler
 from tqdm import tqdm
 
-from tensorflow_tts.processor import LJSpeechProcessor, BakerProcessor
+from tensorflow_tts.processor import LJSpeechProcessor
+from tensorflow_tts.processor import BakerProcessor
+
 from tensorflow_tts.utils import remove_outlier
 
 os.environ["CUDA_VISIBLE_DEVICES"] = ""
@@ -60,8 +62,8 @@ def parse_and_config():
         "--dataset",
         type=str,
         default="ljspeech",
-        choices=["ljspeech", "baker"],
-        help="Dataset to preprocess. Currently only LJSpeech.",
+        choices=["ljspeech, baker"],
+        help="Dataset to preprocess. Currently only (LJSpeech, baker)",
     )
     parser.add_argument(
         "--config", type=str, required=True, help="YAML format configuration file."
@@ -252,15 +254,11 @@ def preprocess():
     """Run preprocessing process and compute statistics for normalizing."""
     config = parse_and_config()
 
-    dataset_processor = {
-        "ljspeech": LJSpeechProcessor,
-        "baker": BakerProcessor
-    }
+    dataset_processor = {"ljspeech": LJSpeechProcessor, "baker": BakerProcessor}
 
     logging.info(f"Selected '{config['dataset']}' processor.")
     processor = dataset_processor[config["dataset"]](
-        config["rootdir"], cleaner_names="english_cleaners",
-        target_rate=config["sampling_rate"]
+        config["rootdir"], cleaner_names="english_cleaners"
     )
 
     # check output directories
@@ -289,9 +287,7 @@ def preprocess():
     # define map iterator
     def iterator_data(items_list):
         for item in items_list:
-            sample = processor.get_one_sample(item)
-            if sample is not None:
-                yield sample
+            yield processor.get_one_sample(item)
 
     train_iterator_data = iterator_data(train_split)
     valid_iterator_data = iterator_data(valid_split)
@@ -417,4 +413,5 @@ def compute_statistics():
     # save statistics to file
     logging.info("Saving computed statistics.")
     scaler_list = [(scaler_mel, ""), (scaler_energy, "_energy"), (scaler_f0, "_f0")]
-    save_statistics_to_file(scaler_list, config)
+    save_statistics_to_file(scaler_list, config)
+

tensorflow_tts/processor/baker.py

@@ -40,7 +40,7 @@ def process_phonelabel(label_file):
 
 class BakerProcessor(object):
 
-    def __init__(self, data_dir, target_rate=None, cleaner_names=None):
+    def __init__(self, data_dir, target_rate=24000, cleaner_names=None):
         self.root_path = data_dir
         self.target_rate = target_rate
 

tensorflow_tts/processor/ljspeech.py

@@ -122,7 +122,7 @@
 class LJSpeechProcessor(object):
     """LJSpeech processor."""
 
-    def __init__(self, data_dir, cleaner_names, target_rate=22050, metadata_filename="metadata.csv"):
+    def __init__(self, data_dir, cleaner_names, metadata_filename="metadata.csv"):
         self.data_dir = data_dir
         self.cleaner_names = cleaner_names