sync with applio

rvc/infer/infer.py

@@ -23,6 +23,7 @@
 
 logging.getLogger("httpx").setLevel(logging.WARNING)
 logging.getLogger("httpcore").setLevel(logging.WARNING)
+logging.getLogger("faiss").setLevel(logging.WARNING)
 
 
 class VoiceConverter:
@@ -79,7 +80,7 @@ def remove_audio_noise(input_audio_path, reduction_strength=0.7):
  )
  return reduced_noise
  except Exception as error:
- print(f"Error cleaning audio: {error}")
+ print(f"An error occurred removing audio noise: {error}")
  return None
 
  @staticmethod
@@ -114,7 +115,7 @@ def convert_audio_format(input_path, output_path, output_format):
  sf.write(output_path, audio, target_sr, format=output_format.lower())
  return output_path
  except Exception as error:
- print(f"Failed to convert audio to {output_format} format: {error}")
+ print(f"An error occurred converting the audio format: {error}")
 
  def convert_audio(
  self,
@@ -233,15 +234,15 @@ def convert_audio(
  split_audio=False,
  f0_autotune=f0_autotune,
  filter_radius=filter_radius,
+ export_format=export_format,
+ upscale_audio=upscale_audio,
  embedder_model=embedder_model,
  embedder_model_custom=embedder_model_custom,
  clean_audio=clean_audio,
  clean_strength=clean_strength,
- export_format=export_format,
- upscale_audio=upscale_audio,
  )
  except Exception as error:
- print(f"Error in processing split audio segment: {error}")
+ print(f"An error occurred processing the segmented audio: {error}")
  print(traceback.format_exc())
  return f"Error {error}"
  print("Finished processing segmented audio, now merging audio...")
@@ -299,7 +300,7 @@ def convert_audio(
  )
 
  except Exception as error:
- print(f"Voice conversion failed: {error}")
+ print(f"An error occurred during audio conversion: {error}")
  print(traceback.format_exc())
 
  def get_vc(self, weight_root, sid):

rvc/infer/pipeline.py

@@ -9,14 +9,17 @@
 import librosa
 import numpy as np
 from scipy import signal
-from functools import lru_cache
 from torch import Tensor
 
 now_dir = os.getcwd()
 sys.path.append(now_dir)
+
 from rvc.lib.predictors.RMVPE import RMVPE0Predictor
 from rvc.lib.predictors.FCPE import FCPEF0Predictor
 
+import logging
+
+logging.getLogger("faiss").setLevel(logging.WARNING)
 
 # Constants for high-pass filter
 FILTER_ORDER = 5
@@ -541,7 +544,7 @@ def pipeline(
  index = faiss.read_index(file_index)
  big_npy = index.reconstruct_n(0, index.ntotal)
  except Exception as error:
- print(error)
+ print(f"An error occurred reading the FAISS index: {error}")
  index = big_npy = None
  else:
  index = big_npy = None
@@ -576,7 +579,7 @@ def pipeline(
  inp_f0.append([float(i) for i in line.split(",")])
  inp_f0 = np.array(inp_f0, dtype="float32")
  except Exception as error:
- print(error)
+ print(f"An error occurred reading the F0 file: {error}")
  sid = torch.tensor(sid, device=self.device).unsqueeze(0).long()
  if pitch_guidance == True:
  pitch, pitchf = self.get_f0(

rvc/lib/predictors/FCPE.py

@@ -26,7 +26,7 @@ def load_wav_to_torch(full_path, target_sr=None, return_empty_on_exception=False
  try:
  data, sampling_rate = sf.read(full_path, always_2d=True) # than soundfile.
  except Exception as error:
- print(f"'{full_path}' failed to load with {error}")
+ print(f"An error occurred loading {full_path}: {error}")
  if return_empty_on_exception:
  return [], sampling_rate or target_sr or 48000
  else:

rvc/lib/tools/model_download.py

@@ -76,7 +76,9 @@ def download_from_url(url):
  fuzzy=True,
  )
  except Exception as error:
- error_message = str(error)
+ error_message = str(
+ f"An error occurred downloading the file: {error}"
+ )
  if (
  "Too many users have viewed or downloaded this file recently"
  in error_message
@@ -132,8 +134,8 @@ def download_from_url(url):
  else:
  os.chdir(file_path)
  return None
- except Exception as e:
- print(e)
+ except Exception as error:
+ print(f"An error occurred downloading the file: {error}")
  os.chdir(file_path)
  return None
 
@@ -238,7 +240,7 @@ def download_from_url(url):
  wget.download(url)
  except Exception as error:
  os.chdir(now_dir)
- print(error)
+ print(f"An error occurred downloading the file: {error}")
  return None
 
  for currentPath, _, zipFiles in os.walk(zips_path):
@@ -265,7 +267,7 @@ def extract_and_show_progress(zipfile_path, unzips_path):
  os.remove(zipfile_path)
  return True
  except Exception as error:
- print(error)
+ print(f"An error occurred extracting the zip file: {error}")
  return False
 
 

rvc/lib/tools/prerequisites_download.py

@@ -137,7 +137,7 @@ def prequisites_download_pipeline(pretraineds_v1, pretraineds_v2, models, exe):
  download_mapping_files(pretraineds_v2_list)
 
  # Clear the console after all downloads are completed
- clear_console() 
+ clear_console()
 
 
 def clear_console():

rvc/lib/tools/split_audio.py

@@ -52,8 +52,8 @@ def process_audio(file_path):
 
  return "Finish", new_dir_path
 
- except Exception as e:
- print(f"An error occurred: {e}")
+ except Exception as error:
+ print(f"An error occurred splitting the audio: {error}")
  return "Error", None
 
 
@@ -103,5 +103,5 @@ def merge_audio(timestamps_file):
  # print(f"Exported merged file: {merged_filename}\n")
  return merged_audio.frame_rate, merged_audio_np
 
- except Exception as e:
- print(f"An error occurred: {e}")
+ except Exception as error:
+ print(f"An error occurred splitting the audio: {error}")

rvc/lib/utils.py

@@ -24,7 +24,7 @@ def load_audio(file, sample_rate):
  if sr != sample_rate:
  audio = librosa.resample(audio, orig_sr=sr, target_sr=sample_rate)
  except Exception as error:
- raise RuntimeError(f"Failed to load audio: {error}")
+ raise RuntimeError(f"An error occurred loading the audio: {error}")
 
  return audio.flatten()
 

rvc/train/data_utils.py

@@ -118,9 +118,7 @@ def get_audio(self, filename):
  audio, sample_rate = load_wav_to_torch(filename)
  if sample_rate != self.sample_rate:
  raise ValueError(
- "{} SR doesn't match target {} SR".format(
- sample_rate, self.sample_rate
- )
+ "{} SR doesn't match target {} SR".format(sample_rate, self.sample_rate)
  )
  audio_norm = audio
  audio_norm = audio_norm.unsqueeze(0)
@@ -129,7 +127,7 @@ def get_audio(self, filename):
  try:
  spec = torch.load(spec_filename)
  except Exception as error:
- print(f"{spec_filename}: {error}")
+ print(f"An error occurred getting spec from {spec_filename}: {error}")
  spec = spectrogram_torch(
  audio_norm,
  self.filter_length,
@@ -344,9 +342,7 @@ def get_audio(self, filename):
  audio, sample_rate = load_wav_to_torch(filename)
  if sample_rate != self.sample_rate:
  raise ValueError(
- "{} SR doesn't match target {} SR".format(
- sample_rate, self.sample_rate
- )
+ "{} SR doesn't match target {} SR".format(sample_rate, self.sample_rate)
  )
  audio_norm = audio
  audio_norm = audio_norm.unsqueeze(0)
@@ -355,7 +351,7 @@ def get_audio(self, filename):
  try:
  spec = torch.load(spec_filename)
  except Exception as error:
- print(f"{spec_filename}: {error}")
+ print(f"An error occurred getting spec from {spec_filename}: {error}")
  spec = spectrogram_torch(
  audio_norm,
  self.filter_length,

rvc/train/extract/extract_f0_print.py

@@ -105,7 +105,7 @@ def process_file(self, file_info):
  coarse_pit = self.coarse_f0(feature_pit)
  np.save(opt_path1, coarse_pit, allow_pickle=False)
  except Exception as error:
- print(f"F0 extraction failed for {inp_path}: {error}")
+ print(f"An error occurred extracting file {inp_path}: {error}")
 
 
 if __name__ == "__main__":
@@ -124,7 +124,7 @@ def process_file(self, file_info):
  input_path = os.path.join(input_root, name)
  output_path1 = os.path.join(output_root1, name)
  output_path2 = os.path.join(output_root2, name)
- np_arr = load_audio(input_path, 16000) #self.fs?
+ np_arr = load_audio(input_path, 16000)  # self.fs?
  paths.append([input_path, output_path1, output_path2, np_arr])
 
  print(f"Starting extraction with {num_processes} cores and {f0_method}...")

rvc/train/extract/extract_feature_print.py

@@ -99,7 +99,7 @@ def read_wave(wav_path, normalize=False):
  print(f"{file} is invalid")
  pbar.set_description(f"Processing {file} {feats.shape}")
  except Exception as error:
- print(error)
+ print(f"An error occurred processing {file}: {error}")
  pbar.update(1)
 
  elapsed_time = time.time() - start_time

rvc/train/extract/preparing_files.py

@@ -15,7 +15,9 @@ def generate_config(rvc_version: str, sample_rate: int, model_path: str):
  shutil.copyfile(config_path, config_save_path)
 
 
-def generate_filelist(pitch_guidance: bool, model_path: str, rvc_version: str, sample_rate: int):
+def generate_filelist(
+ pitch_guidance: bool, model_path: str, rvc_version: str, sample_rate: int
+):
  gt_wavs_dir = f"{model_path}/sliced_audios"
  feature_dir = (
  f"{model_path}/v1_extracted"

rvc/train/preprocess/preprocess.py

@@ -99,7 +99,7 @@ def process_audio(self, path: str, idx0: int):
  break
  self.normalize_and_write(tmp_audio, idx0, idx1)
  except Exception as error:
- print(f"{path}: {error}")
+ print(f"An error occurred on {path} path: {error}")
 
  def process_audio_multiprocessing(self, infos: List[Tuple[str, int]]):
  for path, idx0 in infos:
@@ -119,7 +119,7 @@ def process_audio_multiprocessing_input_directory(
  [infos[i::num_processes] for i in range(num_processes)],
  )
  except Exception as error:
- print(error)
+ print(f"An error occurred on {input_root} path: {error}")
 
 
 def preprocess_training_set(

rvc/train/process/change_info.py

@@ -11,4 +11,4 @@ def change_info(path, info, name):
  torch.save(ckpt, f"logs/{name}/{name}")
  return "Success."
  except Exception as error:
- print(error)
+ print(f"An error occurred changing the info: {error}")

rvc/train/process/extract_index.py

@@ -76,8 +76,7 @@
  print(f"Saved index file '{index_filepath_added}'")
 
 except Exception as error:
- print(f"Failed to train index: {error}")
- if "one array to concatenate" in str(error):
- print(
- "If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file."
- )
+ print(f"An error occurred extracting the index: {error}")
+ print(
+ "If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file."
+ )

rvc/train/process/extract_model.py

@@ -21,10 +21,15 @@ def replace_keys_in_dict(d, old_key_part, new_key_part):
 def extract_model(ckpt, sr, pitch_guidance, name, model_dir, epoch, step, version, hps):
  try:
  print(f"Saved model '{model_dir}' (epoch {epoch} and step {step})")
+
+ model_dir_path = os.path.dirname(model_dir)
+ os.makedirs(model_dir_path, exist_ok=True)
+
  pth_file = f"{name}_{epoch}e_{step}s.pth"
  pth_file_old_version_path = os.path.join(
- model_dir, f"{pth_file}_old_version.pth"
+ model_dir_path, f"{pth_file}_old_version.pth"
  )
+
  opt = OrderedDict(
  weight={
  key: value.half() for key, value in ckpt.items() if "enc_q" not in key
@@ -62,7 +67,7 @@ def extract_model(ckpt, sr, pitch_guidance, name, model_dir, epoch, step, versio
  model_hash = hashlib.sha256(hash_input.encode()).hexdigest()
  opt["model_hash"] = model_hash
 
- torch.save(opt, model_dir)
+ torch.save(opt, os.path.join(model_dir_path, pth_file))
 
  model = torch.load(model_dir, map_location=torch.device("cpu"))
  torch.save(
@@ -79,4 +84,4 @@ def extract_model(ckpt, sr, pitch_guidance, name, model_dir, epoch, step, versio
  os.rename(pth_file_old_version_path, model_dir)
 
  except Exception as error:
- print(error)
+ print(f"An error occurred extracting the model: {error}")

rvc/train/process/extract_small_model.py

@@ -21,7 +21,15 @@ def replace_keys_in_dict(d, old_key_part, new_key_part):
  return updated_dict
 
 
-def extract_small_model(path: str, name: str, sr: int, pitch_guidance: bool, version: str, epoch: int, step: int):
+def extract_small_model(
+ path: str,
+ name: str,
+ sr: int,
+ pitch_guidance: bool,
+ version: str,
+ epoch: int,
+ step: int,
+):
  try:
  ckpt = torch.load(path, map_location="cpu")
  pth_file = f"{name}.pth"
@@ -172,4 +180,4 @@ def extract_small_model(path: str, name: str, sr: int, pitch_guidance: bool, ver
  os.remove(pth_file_old_version_path)
  os.rename(pth_file_old_version_path, pth_file)
  except Exception as error:
- print(error)
+ print(f"An error occurred extracting the model: {error}")

rvc/train/process/model_blender.py

@@ -2,6 +2,7 @@
 import torch
 from collections import OrderedDict
 
+
 def extract(ckpt):
  a = ckpt["model"]
  opt = OrderedDict()
@@ -12,15 +13,16 @@ def extract(ckpt):
  opt["weight"][key] = a[key]
  return opt
 
+
 def model_blender(name, path1, path2, ratio):
  try:
  message = f"Model {path1} and {path2} are merged with alpha {ratio}."
  ckpt1 = torch.load(path1, map_location="cpu")
  ckpt2 = torch.load(path2, map_location="cpu")
- 
+
  if ckpt1["sr"] != ckpt2["sr"]:
  return "The sample rates of the two models are not the same."
- 
+
  cfg = ckpt1["config"]
  cfg_f0 = ckpt1["f0"]
  cfg_version = ckpt1["version"]
@@ -61,5 +63,5 @@ def model_blender(name, path1, path2, ratio):
  print(message)
  return message, os.path.join("logs", "%s.pth" % name)
  except Exception as error:
- print(error)
+ print(f"An error occurred blending the models: {error}")
  return error