[fast_inference] 回退策略，减少padding影响，开放选项，同步代码 (#986)

* Update README

* Optimize-English-G2P

* docs: change akward expression

* docs: update Changelog_KO.md

* Fix CN punc in EN,add 's match

* Adjust normalize and g2p logic

* Update zh_CN.json

* Update README (#827)

Update README.md
Update some outdated file paths and commands

* 修复英文多音字,调整字典热加载,新增姓名匹配 (#869)

* Fix homograph dict

* Add JSON in dict

* Adjust hot dict to hot reload

* Add English name dict

* Adjust get name dict logic

* Make API Great Again (#894)

* Add zh/jp/en mix

* Optimize code readability and formatted output.

* Try OGG streaming

* Add stream mode arg

* Add media type arg

* Add cut punc arg

* Eliminate punc risk

* Update README (#895)

* Update README

* Update README

* update README

* update README

* fix typo s/Licence /License (#904)

* fix reformat cmd (#917)

Co-authored-by: starylan <starylan@outlook.com>

* Update README.md

* Normalize chinese arithmetic operations (#947)

* 改变训练和推理时的mask策略，以修复当batch_size>1时，产生的复读现象

* 同步main分支代码，增加“保持随机”选项

* 在colab中运行colab_webui.ipynb发生的uvr5模型缺失问题 (#968)

在colab中使用git下载uvr5模型时报错：
fatal: destination path 'uvr5_weights' already exists and is not an empty directory.
通过在下载前将原本从本仓库下载的uvr5_weights文件夹删除可以解决问题。

* [ASR] 修复FasterWhisper遍历输入路径失败 (#956)

* remove glob

* rename

* reset mirror pos

* 回退mask策略；
回退pad策略；
在T2SBlock中添加padding_mask，以减少pad的影响；
开放repetition_penalty参数，让用户自行调整重复惩罚的强度；
增加parallel_infer参数，用于开启或关闭并行推理，关闭时与0307版本保持一致；
在webui中增加“保持随机”选项；
同步main分支代码。

* 删除无用注释

---------

Co-authored-by: Lion <drain.daters.0p@icloud.com>
Co-authored-by: RVC-Boss <129054828+RVC-Boss@users.noreply.github.com>
Co-authored-by: KamioRinn <snowsdream@live.com>
Co-authored-by: Pengoose <pengoose_dev@naver.com>
Co-authored-by: Yuan-Man <68322456+Yuan-ManX@users.noreply.github.com>
Co-authored-by: XXXXRT666 <157766680+XXXXRT666@users.noreply.github.com>
Co-authored-by: KamioRinn <63162909+KamioRinn@users.noreply.github.com>
Co-authored-by: Lion-Wu <130235128+Lion-Wu@users.noreply.github.com>
Co-authored-by: digger yu <digger-yu@outlook.com>
Co-authored-by: SapphireLab <36986837+SapphireLab@users.noreply.github.com>
Co-authored-by: starylan <starylan@outlook.com>
Co-authored-by: shadow01a <141255649+shadow01a@users.noreply.github.com>

tools/asr/fasterwhisper_asr.py

@@ -1,18 +1,16 @@
 import argparse
 import os
-os.environ["HF_ENDPOINT"]="https://hf-mirror.com"
 import traceback
-import requests
-from glob import glob
-import torch
 
+os.environ["HF_ENDPOINT"]          = "https://hf-mirror.com"
+os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"
+
+import torch
 from faster_whisper import WhisperModel
 from tqdm import tqdm
 
 from tools.asr.config import check_fw_local_models
 
-os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
-
 language_code_list = [
     "af", "am", "ar", "as", "az", 
     "ba", "be", "bg", "bn", "bo", 
@@ -36,7 +34,7 @@ language_code_list = [
     "vi", "yi", "yo", "zh", "yue",
     "auto"]
 
-def execute_asr(input_folder, output_folder, model_size, language,precision):
+def execute_asr(input_folder, output_folder, model_size, language, precision):
     if '-local' in model_size:
         model_size = model_size[:-6]
         model_path = f'tools/asr/models/faster-whisper-{model_size}'
@@ -50,17 +48,18 @@ def execute_asr(input_folder, output_folder, model_size, language,precision):
         model = WhisperModel(model_path, device=device, compute_type=precision)
     except:
         return print(traceback.format_exc())
+    
+    input_file_names = os.listdir(input_folder)
+    input_file_names.sort()
+
     output = []
     output_file_name = os.path.basename(input_folder)
-    output_file_path = os.path.abspath(f'{output_folder}/{output_file_name}.list')
-
-    if not os.path.exists(output_folder):
-        os.makedirs(output_folder)
-
-    for file in tqdm(glob(os.path.join(input_folder, '**/*.wav'), recursive=True)):
+    
+    for file_name in tqdm(input_file_names):
         try:
+            file_path = os.path.join(input_folder, file_name)
             segments, info = model.transcribe(
-                audio          = file,
+                audio          = file_path,
                 beam_size      = 5,
                 vad_filter     = True,
                 vad_parameters = dict(min_silence_duration_ms=700),
@@ -68,18 +67,23 @@ def execute_asr(input_folder, output_folder, model_size, language,precision):
             text = ''
 
             if info.language == "zh":
-                print("检测为中文文本,转funasr处理")
+                print("检测为中文文本, 转 FunASR 处理")
                 if("only_asr"not in globals()):
-                    from tools.asr.funasr_asr import only_asr##如果用英文就不需要导入下载模型
-                text = only_asr(file)
+                    from tools.asr.funasr_asr import \
+                        only_asr  # #如果用英文就不需要导入下载模型
+                text = only_asr(file_path)
 
             if text == '':
                 for segment in segments:
                     text += segment.text
-            output.append(f"{file}|{output_file_name}|{info.language.upper()}|{text}")
+            output.append(f"{file_path}|{output_file_name}|{info.language.upper()}|{text}")
         except:
             return print(traceback.format_exc())
-        
+    
+    output_folder = output_folder or "output/asr_opt"
+    os.makedirs(output_folder, exist_ok=True)
+    output_file_path = os.path.abspath(f'{output_folder}/{output_file_name}.list')
+
     with open(output_file_path, "w", encoding="utf-8") as f:
         f.write("\n".join(output))
         print(f"ASR 任务完成->标注文件路径: {output_file_path}\n")

tools/asr/funasr_asr.py

@@ -38,10 +38,11 @@ def execute_asr(input_folder, output_folder, model_size, language):
     output = []
     output_file_name = os.path.basename(input_folder)
 
-    for name in tqdm(input_file_names):
+    for file_name in tqdm(input_file_names):
         try:
-            text = model.generate(input="%s/%s"%(input_folder, name))[0]["text"]
-            output.append(f"{input_folder}/{name}|{output_file_name}|{language.upper()}|{text}")
+            file_path = os.path.join(input_folder, file_name)
+            text = model.generate(input=file_path)[0]["text"]
+            output.append(f"{file_path}|{output_file_name}|{language.upper()}|{text}")
         except:
             print(traceback.format_exc())
 

tools/uvr5/webui.py

@@ -73,8 +73,7 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
                     os.path.basename(inp_path),
                 )
                 os.system(
-                    "ffmpeg -i %s -vn -acodec pcm_s16le -ac 2 -ar 44100 %s -y"
-                    % (inp_path, tmp_path)
+                    f'ffmpeg -i "{inp_path}" -vn -acodec pcm_s16le -ac 2 -ar 44100 "{tmp_path}" -y'
                 )
                 inp_path = tmp_path
             try: