add ASR_Model Select

tools/asr/config.py

@@ -0,0 +1,31 @@
+import os
+
+def check_fw_local_models():
+    '''
+    启动时检查本地是否有 Faster Whisper 模型.
+    '''
+    model_size_list = [
+        "tiny",     "tiny.en", 
+        "base",     "base.en", 
+        "small",    "small.en", 
+        "medium",   "medium.en", 
+        "large",    "large-v1", 
+        "large-v2", "large-v3"]
+    for i, size in enumerate(model_size_list):
+        if os.path.exists(f'tools/asr/models/faster-whisper-{size}'):
+            model_size_list[i] = size + '(local)'
+    return model_size_list
+
+asr_dict = {
+    "达摩 ASR (中文)": {
+        'lang': ['zh'],
+        'size': ['large'],
+        'path': 'funasr_asr.py',
+    },
+    "Faster Whisper (多语种)": {
+        'lang': ['auto', 'zh', 'en', 'ja'],
+        'size': check_fw_local_models(),
+        'path': 'fasterwhisper_asr.py'
+    }
+}
+

tools/asr/fasterwhisper_asr.py

@@ -0,0 +1,97 @@
+import argparse
+import os
+import traceback
+import requests
+from glob import glob
+
+from faster_whisper import WhisperModel
+from tqdm import tqdm
+
+from 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", 
+    "br", "bs", "ca", "cs", "cy", 
+    "da", "de", "el", "en", "es", 
+    "et", "eu", "fa", "fi", "fo", 
+    "fr", "gl", "gu", "ha", "haw", 
+    "he", "hi", "hr", "ht", "hu", 
+    "hy", "id", "is", "it", "ja", 
+    "jw", "ka", "kk", "km", "kn", 
+    "ko", "la", "lb", "ln", "lo", 
+    "lt", "lv", "mg", "mi", "mk", 
+    "ml", "mn", "mr", "ms", "mt", 
+    "my", "ne", "nl", "nn", "no", 
+    "oc", "pa", "pl", "ps", "pt", 
+    "ro", "ru", "sa", "sd", "si", 
+    "sk", "sl", "sn", "so", "sq", 
+    "sr", "su", "sv", "sw", "ta", 
+    "te", "tg", "th", "tk", "tl", 
+    "tr", "tt", "uk", "ur", "uz", 
+    "vi", "yi", "yo", "zh", "yue",
+    "auto"]
+
+def execute_asr(input_folder, output_folder, model_size, language):
+    if 'local' in model_size:
+        model_size = model_size.split('(')[0]
+        model_path = f'tools/asr/models/faster-whisper-{model_size}'
+    else:
+        model_path = model_size
+    if language == 'auto':
+        language = None #不设置语种由模型自动输出概率最高的语种
+
+    try:
+        model = WhisperModel(model_path, device="cuda", compute_type="float16")
+    except:
+        return print(traceback.format_exc())
+
+    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)):
+        try:
+            segments, info = model.transcribe(
+                audio          = file,
+                beam_size      = 5,
+                vad_filter     = True,
+                vad_parameters = dict(min_silence_duration_ms=700),
+                language       = language)
+            text = ''
+            for segment in segments:
+                text += segment.text
+            output.append(f"{file}|{output_file_name}|{info.language.upper()}|{text}")
+        except:
+            return print(traceback.format_exc())
+        
+    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")
+    return output_file_path
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-i", "--input_folder", type=str, required=True,
+                        help="Path to the folder containing WAV files.")
+    parser.add_argument("-o", "--output_folder", type=str, required=True, 
+                        help="Output folder to store transcriptions.")
+    parser.add_argument("-s", "--model_size", type=str, default='large-v3', 
+                        choices=check_fw_local_models(),
+                        help="Model Size of Faster Whisper")
+    parser.add_argument("-l", "--language", type=str, default='zh', 
+                        choices=language_code_list,
+                        help="Language of the audio files.")
+
+    cmd = parser.parse_args()
+    output_file_path = execute_asr(
+        input_folder  = cmd.input_folder,
+        output_folder = cmd.output_folder,
+        model_size    = cmd.model_size,
+        language      = cmd.language,
+    )

tools/asr/funasr_asr.py

@@ -0,0 +1,66 @@
+# -*- coding:utf-8 -*-
+
+import argparse
+import os
+import traceback
+from tqdm import tqdm
+
+from funasr import AutoModel
+
+path_asr  = 'tools/damo_asr/models/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'
+path_vad  = 'tools/damo_asr/models/speech_fsmn_vad_zh-cn-16k-common-pytorch'
+path_punc = 'tools/damo_asr/models/punc_ct-transformer_zh-cn-common-vocab272727-pytorch'
+path_asr  = path_asr  if os.path.exists(path_asr)  else "iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
+path_vad  = path_vad  if os.path.exists(path_vad)  else "iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
+path_punc = path_punc if os.path.exists(path_punc) else "iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
+
+model = AutoModel(
+    model               = path_asr,
+    model_revision      = "v2.0.4",
+    vad_model           = path_vad,
+    vad_model_revision  = "v2.0.4",
+    punc_model          = path_punc,
+    punc_model_revision = "v2.0.4",
+)
+
+def execute_asr(input_folder, output_folder, model_size, language):
+    input_file_names = os.listdir(input_folder)
+    input_file_names.sort()
+    
+    output = []
+    output_file_name = os.path.basename(input_folder)
+
+    for 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}")
+        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")
+    return output_file_path
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-i", "--input_folder", type=str, required=True,
+                        help="Path to the folder containing WAV files.")
+    parser.add_argument("-o", "--output_folder", type=str, required=True, 
+                        help="Output folder to store transcriptions.")
+    parser.add_argument("-s", "--model_size", type=str, default='large',
+                        help="Model Size of FunASR is Large")
+    parser.add_argument("-l", "--language", type=str, default='zh', choices=['zh'],
+                        help="Language of the audio files.")
+
+    cmd = parser.parse_args()
+    execute_asr(
+        input_folder  = cmd.input_folder,
+        output_folder = cmd.output_folder,
+        model_size    = cmd.model_size,
+        language      = cmd.language,
+    )

tools/damo_asr/WhisperASR.py

@@ -1,42 +0,0 @@
-import os
-import argparse
-import os
-os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
-from glob import glob
-from faster_whisper import WhisperModel
-
-def main(input_folder, output_folder, output_filename, language):
-    model = WhisperModel("large-v3", device="cuda", compute_type="float16")
-
-    output_file = os.path.join(output_folder, output_filename)
-    if not os.path.exists(output_folder):
-        os.makedirs(output_folder)
-
-    with open(output_file, 'w', encoding='utf-8') as f:
-        for file in glob(os.path.join(input_folder, '**/*.wav'), recursive=True):
-            segments, _ = model.transcribe(file, beam_size=10, vad_filter=True,
-                                           vad_parameters=dict(min_silence_duration_ms=700), language=language)
-            segments = list(segments)
-
-            filename = os.path.basename(file).replace('.wav', '')
-            directory = os.path.dirname(file)
-
-            result_line = f"{file}|{language.upper()}|{segments[0].text}\n"
-            f.write(result_line)
-
-if __name__ == '__main__':
-    parser = argparse.ArgumentParser()
-    parser.add_argument("-i", "--input_folder", type=str, required=True,
-                        help="Path to the folder containing WAV files.")
-    parser.add_argument("-o", "--output_folder", type=str, required=True, help="Output folder to store transcriptions.")
-    parser.add_argument("-f", "--output_filename", type=str, default="transcriptions.txt", help="Name of the output text file.")
-    parser.add_argument("-l", "--language", type=str, default='zh', choices=['zh', 'en', ...],
-                        help="Language of the audio files.")
-
-    cmd = parser.parse_args()
-
-    input_folder = cmd.input_folder
-    output_folder = cmd.output_folder
-    output_filename = cmd.output_filename
-    language = cmd.language
-    main(input_folder, output_folder, output_filename, language)

tools/damo_asr/cmd-asr.py

@@ -1,39 +0,0 @@
-# -*- coding:utf-8 -*-
-
-import sys,os,traceback
-
-from funasr import AutoModel
-
-dir=sys.argv[1]
-if(dir[-1]=="/"):dir=dir[:-1]
-# opt_name=dir.split("\\")[-1].split("/")[-1]
-opt_name=os.path.basename(dir)
-
-path_asr='tools/damo_asr/models/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'
-path_vad='tools/damo_asr/models/speech_fsmn_vad_zh-cn-16k-common-pytorch'
-path_punc='tools/damo_asr/models/punc_ct-transformer_zh-cn-common-vocab272727-pytorch'
-path_asr=path_asr if os.path.exists(path_asr)else "iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
-path_vad=path_vad if os.path.exists(path_vad)else "iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
-path_punc=path_punc if os.path.exists(path_punc)else "iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-
-model = AutoModel(model=path_asr, model_revision="v2.0.4",
-                  vad_model=path_vad,
-                  vad_model_revision="v2.0.4",
-                  punc_model=path_punc,
-                  punc_model_revision="v2.0.4",
-                  )
-
-
-opt=[]
-file_names = os.listdir(dir)
-file_names.sort()
-for name in file_names:
-    try:
-        text = model.generate(input="%s/%s"%(dir,name))[0]["text"]
-        opt.append("%s/%s|%s|ZH|%s"%(dir,name,opt_name,text))
-    except:
-        print(traceback.format_exc())
-
-opt_dir="output/asr_opt"
-os.makedirs(opt_dir,exist_ok=True)
-with open("%s/%s.list"%(opt_dir,opt_name),"w",encoding="utf-8")as f:f.write("\n".join(opt))

webui.py

@@ -192,20 +192,26 @@ def change_tts_inference(if_tts,bert_path,cnhubert_base_path,gpu_number,gpt_path
         p_tts_inference=None
         yield i18n("TTS推理进程已关闭")
 
-
-def open_asr(asr_inp_dir):
+from tools.asr.config import asr_dict
+def open_asr(asr_inp_dir, asr_opt_dir, asr_model, asr_model_size, asr_lang):
     global p_asr
     if(p_asr==None):
         asr_inp_dir=my_utils.clean_path(asr_inp_dir)
-        cmd = '"%s" tools/damo_asr/cmd-asr.py "%s"'%(python_exec,asr_inp_dir)
+        cmd = f"{python_exec} tools/asr/{asr_dict[asr_model]['path']}"
+        cmd += f" -i {asr_inp_dir}"
+        cmd += f" -o {asr_opt_dir}"
+        cmd += f" -s {asr_model_size}"
+        cmd += f" -l {asr_lang}"                
+
         yield "ASR任务开启：%s"%cmd,{"__type__":"update","visible":False},{"__type__":"update","visible":True}
         print(cmd)
         p_asr = Popen(cmd, shell=True)
         p_asr.wait()
         p_asr=None
-        yield "ASR任务完成",{"__type__":"update","visible":True},{"__type__":"update","visible":False}
+        yield f"ASR任务完成, 查看终端进行下一步",{"__type__":"update","visible":True},{"__type__":"update","visible":False}
     else:
         yield "已有正在进行的ASR任务，需先终止才能开启下一次任务",{"__type__":"update","visible":False},{"__type__":"update","visible":True}
+        return None
 
 def close_asr():
     global p_asr
@@ -674,12 +680,44 @@ with gr.Blocks(title="GPT-SoVITS WebUI") as app:
             with gr.Row():
                 open_asr_button = gr.Button(i18n("开启离线批量ASR"), variant="primary",visible=True)
                 close_asr_button = gr.Button(i18n("终止ASR进程"), variant="primary",visible=False)
-                asr_inp_dir = gr.Textbox(
-                    label=i18n("批量ASR(中文only)输入文件夹路径"),
-                    value="D:\\RVC1006\\GPT-SoVITS\\raw\\xxx",
-                    interactive=True,
-                )
-                asr_info = gr.Textbox(label=i18n("ASR进程输出信息"))
+                with gr.Column():
+                    with gr.Row():
+                        asr_inp_dir = gr.Textbox(
+                            label=i18n("输入文件夹路径"),
+                            value="D:\\RVC1006\\GPT-SoVITS\\raw\\xxx",
+                            interactive=True,
+                        )
+                        asr_opt_dir = gr.Textbox(
+                            label       = i18n("输出文件夹路径"),
+                            value       = "output/asr_opt",
+                            interactive = False,
+                        )
+                    with gr.Row():
+                        asr_model = gr.Dropdown(
+                            label       = i18n("ASR 模型"),
+                            choices     = list(asr_dict.keys()),
+                            interactive = True,
+                        )
+                        asr_size = gr.Dropdown(
+                            label       = i18n("ASR 模型尺寸"),
+                            choices     = [],
+                            interactive = True,
+                        )
+                        asr_lang = gr.Dropdown(
+                            label       = i18n("ASR 语言设置"),
+                            choices     = [],
+                            interactive = True,
+                        )
+                    with gr.Row():
+                        asr_info = gr.Textbox(label=i18n("ASR进程输出信息"))
+
+                def change_lang_choices(key): #根据选择的模型修改可选的语言
+                    return gr.Dropdown(choices=asr_dict[key]['lang'])
+                def change_size_choices(key): # 根据选择的模型修改可选的模型尺寸
+                    return gr.Dropdown(choices=asr_dict[key]['size'])
+                asr_model.change(change_lang_choices, asr_model, asr_lang)
+                asr_model.change(change_size_choices, asr_model, asr_size)
+                
             gr.Markdown(value=i18n("0d-语音文本校对标注工具"))
             with gr.Row():
                 if_label = gr.Checkbox(label=i18n("是否开启打标WebUI"),show_label=True)
@@ -691,7 +729,7 @@ with gr.Blocks(title="GPT-SoVITS WebUI") as app:
                 label_info = gr.Textbox(label=i18n("打标工具进程输出信息"))
             if_label.change(change_label, [if_label,path_list], [label_info])
             if_uvr5.change(change_uvr5, [if_uvr5], [uvr5_info])
-            open_asr_button.click(open_asr, [asr_inp_dir], [asr_info,open_asr_button,close_asr_button])
+            open_asr_button.click(open_asr, [asr_inp_dir, asr_opt_dir, asr_model, asr_size, asr_lang], [asr_info,open_asr_button,close_asr_button])
             close_asr_button.click(close_asr, [], [asr_info,open_asr_button,close_asr_button])
             open_slicer_button.click(open_slice, [slice_inp_path,slice_opt_root,threshold,min_length,min_interval,hop_size,max_sil_kept,_max,alpha,n_process], [slicer_info,open_slicer_button,close_slicer_button])
             close_slicer_button.click(close_slice, [], [slicer_info,open_slicer_button,close_slicer_button])
@@ -788,7 +826,7 @@ with gr.Blocks(title="GPT-SoVITS WebUI") as app:
                     tts_info = gr.Textbox(label=i18n("TTS推理WebUI进程输出信息"))
                     if_tts.change(change_tts_inference, [if_tts,bert_pretrained_dir,cnhubert_base_dir,gpu_number_1C,GPT_dropdown,SoVITS_dropdown], [tts_info])
         with gr.TabItem(i18n("2-GPT-SoVITS-变声")):gr.Markdown(value=i18n("施工中，请静候佳音"))
-    app.queue(concurrency_count=511, max_size=1022).launch(
+    app.queue(max_size=1022).launch(
         server_name="0.0.0.0",
         inbrowser=True,
         share=is_share,