Command line batch processing of RVC and UVR example

batch-example.cmd

@@ -0,0 +1,4 @@
+@echo off
+cd C:\Users\Conda\Desktop\dev\RVC1006Nvidia
+runtime\python.exe batch.py --pycmd runtime\python.exe
+pause

batch-example.py

@@ -0,0 +1,201 @@
+name="YourName" # Output File will be <YourName-FileName>.mp3
+model_name="js-20231220.pth" # model in the path assets/weights/
+file_index="logs/js-20231220/trained_IVF358_Flat_nprobe_1_js-20231220_v2.index" # index file in the path logs/<modelName>/
+save_root_path="C:/Users/Conda/Desktop/AI_data/audio/result"  
+'''
+使用这个脚本是为了批量转换到单个角色的音频，可以自动进行音轨分离和合并。输入文件支持文件夹和单文件。
+Use this script to batch convert audio to a single character, allowing automatic track separation and merging. Input files support folders and single files.
+Create a bat file to run this script.
+@echo off
+cd C:\Users\Conda\Desktop\dev\RVC1006Nvidia
+runtime\python.exe batch-example.py --pycmd runtime\python.exe
+pause
+'''
+import os, sys, shutil, subprocess
+from dotenv import load_dotenv
+import soundfile as sf
+now_dir = os.getcwd()
+sys.path.append(now_dir)
+load_dotenv()
+from infer.modules.vc.modules import VC
+from configs.config import Config
+from infer.modules.uvr5.modules import uvr
+
+# 底下的没必要改
+f0_file="assets/weights/"+model_name
+file_index2=[]
+f0_method="rmvpe" 
+index_rate=0.75 # 检索特征占比, 0-1
+filter_radius=3 #>=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音, 0-7
+resample_sr=0 # 后处理重采样至最终采样率，0为不进行重采样, 0-48000
+rms_mix_rate=0.25 # 输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络,0-1
+protect=0.33 # 保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果, 0-0.5
+
+def mix_audio(path_accompaniment, path_vocal, output_name): # 使用ffmpeg合并伴奏和人声
+    command = [
+        'ffmpeg',
+        '-i', path_accompaniment,
+        '-i', path_vocal,
+        '-filter_complex', '[1:a]volume=1.5[a1];[0:a][a1]amix=inputs=2:duration=first:dropout_transition=3',
+        output_name
+    ]
+    try:
+        subprocess.run(command, check=True)
+        print(f"Audio mixed successfully into {output_name}")
+    except subprocess.CalledProcessError as e:
+        print(f"Error during audio mixing: {e}")
+
+def RVC(input_audio, vc_transform=0, output_audio = "output_audio.wav"):
+    config = Config()
+    vc = VC(config)
+    fn=vc.get_vc(model_name)
+    result = vc.vc_single(0, input_audio, vc_transform,  # 变调(整数, 半音数量, 升八度12降八度-12)
+                          f0_file, f0_method, file_index, file_index2,
+                          index_rate, filter_radius, 
+                          resample_sr, rms_mix_rate, protect)
+
+    print("Inference result:", result)
+    if result and len(result) > 1 and result[1] is not None:
+        audio_data = result[1][1]
+        sample_rate = 40000
+        sf.write(output_audio, audio_data, sample_rate)
+        print(f"Audio data saved to {output_audio}")
+    else:
+        print("Inference result:", result)
+
+def splitTrack(music_path, save_root_vocal, save_root_ins):
+    model_name="HP5_only_main_vocal"
+    dir_wav_input=""
+    agg=10 #人声提取激进程度, 0-20
+    format0="flac" # 导出文件格式, ["wav", "flac", "mp3", "m4a"]
+    paths=[]
+    inp_root=music_path
+    generator = uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format0)
+    # 迭代生成器以获取并处理每个信息
+    for result in generator:
+        print(result)
+    return True
+
+def find_files(directory, extensions):
+    """Traverse a directory and find files with the given extensions."""
+    files_found = []
+    for root, dirs, files in os.walk(directory):
+        for file in files:
+            if file.split('.')[-1].lower() in extensions:
+                files_found.append(os.path.join(root, file))
+    return files_found
+
+def extract_filename_without_extension(file_path):
+    """
+    Extract the filename without its extension from a full file path.
+
+    Args:
+    file_path (str): The full path to the file.
+
+    Returns:
+    str: The filename without its extension.
+    """
+    filename_with_extension = os.path.basename(file_path)
+    filename_without_extension, _ = os.path.splitext(filename_with_extension)
+    return filename_without_extension
+
+def create_directory_if_not_exists(directory_path):
+    """
+    Create a directory if it does not exist.
+
+    Args:
+    directory_path (str): The path to the directory to create.
+    """
+    if not os.path.exists(directory_path):
+        os.makedirs(directory_path)
+        print(f"Directory created: {directory_path}")
+    else:
+        print(f"Directory already exists: {directory_path}")
+
+def delete_directory(directory_path):
+    try:
+        shutil.rmtree(directory_path)
+        print(f"Directory '{directory_path}' and all its contents have been deleted.")
+    except OSError as error:
+        print(f"Error: {error}. Failed to delete '{directory_path}'.")
+
+def main(input_audio,tone=0):
+    global isMusic, save_path
+    if isMusic:
+        save_path=os.path.join(save_root_path,"music/")
+        music_name=extract_filename_without_extension(input_audio)
+        save_path=os.path.join(save_path,music_name)
+        source_music_path=os.path.join(save_path,"source/")
+        tmp_path=os.path.join(save_path,"tmp")
+        source_music=os.path.join(source_music_path,os.path.basename(input_audio))
+        result_path=os.path.join(save_path,"result/")
+        vocal_path=os.path.join(save_path,"vocal/")
+        result_vocal_path=os.path.join(vocal_path,f"{name}_vocal_{music_name}.wav")
+        final_name=h=os.path.join(result_path,f"{name}_{music_name}.mp3")
+        create_directory_if_not_exists(save_path)
+        create_directory_if_not_exists(input_audio)
+        create_directory_if_not_exists(source_music_path)
+        create_directory_if_not_exists(tmp_path)
+        create_directory_if_not_exists(vocal_path)
+        create_directory_if_not_exists(result_path)
+        print(f"""音乐文件模式\n输入文件：{input_audio}, 保存文件根目录: {save_path},
+复制原音乐保存到：{source_music_path}, 待分离音轨的音乐: {source_music}, 
+最终文件输出目录: {result_path}, result_vocal_path: {result_vocal_path}, final_name: {final_name}""")
+        shutil.copy(input_audio, source_music_path)
+        shutil.copy(input_audio, tmp_path)
+        path_vocal=""; path_instrument=""
+        for i in os.listdir(source_music_path):
+            if "vocal" in i:
+                path_vocal=os.path.join(source_music_path, i)
+            if "instrument" in i:
+                path_instrument=os.path.join(source_music_path, i)
+        if path_vocal!="" and path_instrument!="":
+            print(f"分离后的音轨文件已经存在: 人声路径: {path_vocal}, 伴奏路径: {path_instrument}")
+        else:
+            splitTrack(music_path=tmp_path, save_root_vocal=source_music_path, save_root_ins=source_music_path)
+            for i in os.listdir(source_music_path):
+                if "vocal" in i:
+                    path_vocal=os.path.join(source_music_path, i)
+                if "instrument" in i:
+                    path_instrument=os.path.join(source_music_path, i)
+            print(f"音轨分离完成: 人声路径: {path_vocal}, 伴奏路径: {path_instrument}")
+        RVC(path_vocal, vc_transform=tone, output_audio = result_vocal_path)
+        delete_directory(tmp_path)
+        mix_audio(path_accompaniment=path_instrument, path_vocal=result_vocal_path, output_name=final_name)
+    else:
+        save_path=os.path.join(save_root_path,"basic",name)
+        save_audio_path=os.path.join(save_path,extract_filename_without_extension(input_audio)+".wav")
+        create_directory_if_not_exists(save_path)
+        print(save_audio_path)
+        RVC(input_audio, vc_transform=tone, output_audio = save_audio_path)
+
+while True:
+    input_audio=input("你想处理的文件或者是文件夹(注意如果是目录的话会遍历子目录)：")
+    if os.path.exists(input_audio):
+        break
+    else:
+        print(f"文件或文件夹不存在: {input_audio}")
+while True:
+    tone=input("变调(整数, 半音数量, 升八度12降八度-12), 必须为整数, 默认为0，留空为默认: ")
+    if tone.replace(" ","")=="":
+        tone=0
+        break
+    else:
+        try:
+            tone=int(tone)
+            break
+        except Exception as e:
+            print(f"变调参数错误：{e}")
+            pass
+isMusic=input("是不是音乐文件\n(True 的情况下，会在<save_path>/music文件夹下创建<音乐名>文件夹，自动分离音轨，自动转换，自动合成\nFalse 的情况下，文件会保存到<save_path>/<basic>/<name>文件夹下\n留空默认为False, 即不是音乐\n:")
+if isMusic.replace(" ","")=="":
+    isMusic==False
+else:
+    isMusic==True
+if os.path.isdir(input_audio)==True:
+    files=find_files(directory=input_audio, extensions= ["wav", "flac", "mp3", "m4a"])
+    for i in files:
+        print(i)
+        main(i,tone)
+else:
+    main(input_audio,tone)