Merge pull request #83 from svc-develop-team/optimize-some-code

删除了一些无意义代码

data_utils.py

@@ -47,6 +47,8 @@ class TextAudioSpeakerLoader(torch.utils.data.Dataset):
         audio_norm = audio / self.max_wav_value
         audio_norm = audio_norm.unsqueeze(0)
         spec_filename = filename.replace(".wav", ".spec.pt")
+
+        # Ideally, all data generated after Mar 25 should have .spec.pt
         if os.path.exists(spec_filename):
             spec = torch.load(spec_filename)
         else:

preprocess_flist_config.py

@@ -25,13 +25,11 @@ if __name__ == "__main__":
     parser = argparse.ArgumentParser()
     parser.add_argument("--train_list", type=str, default="./filelists/train.txt", help="path to train list")
     parser.add_argument("--val_list", type=str, default="./filelists/val.txt", help="path to val list")
-    parser.add_argument("--test_list", type=str, default="./filelists/test.txt", help="path to test list")
     parser.add_argument("--source_dir", type=str, default="./dataset/44k", help="path to source dir")
     args = parser.parse_args()
     
     train = []
     val = []
-    test = []
     idx = 0
     spk_dict = {}
     spk_id = 0
@@ -51,13 +49,11 @@ if __name__ == "__main__":
             new_wavs.append(file)
         wavs = new_wavs
         shuffle(wavs)
-        train += wavs[2:-2]
+        train += wavs[2:]
         val += wavs[:2]
-        test += wavs[-2:]
 
     shuffle(train)
     shuffle(val)
-    shuffle(test)
             
     print("Writing", args.train_list)
     with open(args.train_list, "w") as f:
@@ -70,12 +66,6 @@ if __name__ == "__main__":
         for fname in tqdm(val):
             wavpath = fname
             f.write(wavpath + "\n")
-            
-    print("Writing", args.test_list)
-    with open(args.test_list, "w") as f:
-        for fname in tqdm(test):
-            wavpath = fname
-            f.write(wavpath + "\n")
 
     config_template["spk"] = spk_dict
     config_template["model"]["n_speakers"] = spk_id

preprocess_hubert_f0.py

@@ -7,10 +7,12 @@ from random import shuffle
 import torch
 from glob import glob
 from tqdm import tqdm
+from modules.mel_processing import spectrogram_torch
 
 import utils
 import logging
-logging.getLogger('numba').setLevel(logging.WARNING)
+
+logging.getLogger("numba").setLevel(logging.WARNING)
 import librosa
 import numpy as np
 
@@ -29,11 +31,42 @@ def process_one(filename, hmodel):
         wav16k = torch.from_numpy(wav16k).to(device)
         c = utils.get_hubert_content(hmodel, wav_16k_tensor=wav16k)
         torch.save(c.cpu(), soft_path)
+
     f0_path = filename + ".f0.npy"
     if not os.path.exists(f0_path):
-        f0 = utils.compute_f0_dio(wav, sampling_rate=sampling_rate, hop_length=hop_length)
+        f0 = utils.compute_f0_dio(
+            wav, sampling_rate=sampling_rate, hop_length=hop_length
+        )
         np.save(f0_path, f0)
 
+    spec_path = filename.replace(".wav", ".spec.pt")
+    if not os.path.exists(spec_path):
+        # Process spectrogram
+        # The following code can't be replaced by torch.FloatTensor(wav)
+        # because load_wav_to_torch return a tensor that need to be normalized
+
+        audio, sr = utils.load_wav_to_torch(filename)
+        if sr != hps.data.sampling_rate:
+            raise ValueError(
+                "{} SR doesn't match target {} SR".format(
+                    sr, hps.data.sampling_rate
+                )
+            )
+
+        audio_norm = audio / hps.data.max_wav_value
+        audio_norm = audio_norm.unsqueeze(0)
+
+        spec = spectrogram_torch(
+            audio_norm,
+            hps.data.filter_length,
+            hps.data.sampling_rate,
+            hps.data.hop_length,
+            hps.data.win_length,
+            center=False,
+        )
+        spec = torch.squeeze(spec, 0)
+        torch.save(spec, spec_path)
+
 
 def process_batch(filenames):
     print("Loading hubert for content...")
@@ -46,17 +79,23 @@ def process_batch(filenames):
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser()
-    parser.add_argument("--in_dir", type=str, default="dataset/44k", help="path to input dir")
+    parser.add_argument(
+        "--in_dir", type=str, default="dataset/44k", help="path to input dir"
+    )
 
     args = parser.parse_args()
-    filenames = glob(f'{args.in_dir}/*/*.wav', recursive=True)  # [:10]
+    filenames = glob(f"{args.in_dir}/*/*.wav", recursive=True)  # [:10]
     shuffle(filenames)
-    multiprocessing.set_start_method('spawn',force=True)
+    multiprocessing.set_start_method("spawn", force=True)
 
     num_processes = 1
     chunk_size = int(math.ceil(len(filenames) / num_processes))
-    chunks = [filenames[i:i + chunk_size] for i in range(0, len(filenames), chunk_size)]
+    chunks = [
+        filenames[i : i + chunk_size] for i in range(0, len(filenames), chunk_size)
+    ]
     print([len(c) for c in chunks])
-    processes = [multiprocessing.Process(target=process_batch, args=(chunk,)) for chunk in chunks]
+    processes = [
+        multiprocessing.Process(target=process_batch, args=(chunk,)) for chunk in chunks
+    ]
     for p in processes:
         p.start()

spec_gen.py

@@ -1,22 +0,0 @@
-from data_utils import TextAudioSpeakerLoader
-import json
-from tqdm import tqdm
-
-from utils import HParams
-
-config_path = 'configs/config.json'
-with open(config_path, "r") as f:
-    data = f.read()
-config = json.loads(data)
-hps = HParams(**config)
-
-train_dataset = TextAudioSpeakerLoader("filelists/train.txt", hps)
-test_dataset = TextAudioSpeakerLoader("filelists/test.txt", hps)
-eval_dataset = TextAudioSpeakerLoader("filelists/val.txt", hps)
-
-for _ in tqdm(train_dataset):
-    pass
-for _ in tqdm(eval_dataset):
-    pass
-for _ in tqdm(test_dataset):
-    pass