Merge pull request Ikaros-521#344 from Ikaros-521/owner

音频部分相关的优化（音频输出路径自定义等）

.gitignore

@@ -211,5 +211,3 @@ cookie/*.json
 
 # 数据库
 data/data.db
-
-main2.py

config.json

@@ -7,7 +7,8 @@
   "after_prompt": "",
   "comment_log_type": "回答",
   "play_audio": {
-    "enable": true
+    "enable": true,
+    "out_path": "out"
   },
   "show_box": {
     "read_comment": true,

config.json.bak

@@ -7,7 +7,8 @@
   "after_prompt": "",
   "comment_log_type": "回答",
   "play_audio": {
-    "enable": true
+    "enable": true,
+    "out_path": "out"
   },
   "show_box": {
     "read_comment": true,

data/badwords.txt

@@ -2,6 +2,7 @@
 起気
 乞錷
 朲氣
+朲芞
 の朲
 呷朲
 陳希同
@@ -465,7 +466,6 @@
 砍刀
 阳精
 带炼
-鲁迅
 金币网
 李延芝
 开他敏
@@ -1154,7 +1154,6 @@
 办驾驶
 张荣坤
 鄧朴方
-手机
 車侖
 小便
 毛煞逼
@@ -4686,7 +4685,6 @@
 陆肆
 屌七
 山心恨
-断水
 幹死
 淫汁
 小鸡巴

docs/投资人/index.html

@@ -59,7 +59,7 @@
         {
             name: "存在感患者",
             avatar: "https://images.cnblogs.com/cnblogs_com/ikaros-521/2328032/o_230722160559_QQ%E5%9B%BE%E7%89%8720230722234802.jpg",
-            amount: "￥406"
+            amount: "￥410.8"
         },
         {
             name: "月零宫白",
@@ -141,6 +141,11 @@
             avatar: "https://images.cnblogs.com/cnblogs_com/ikaros-521/2328032/o_230925111651_QQ%E5%9B%BE%E7%89%8720230925190851.jpg",
             amount: "￥100"
         },
+        {
+            name: "Jerry",
+            avatar: "https://images.cnblogs.com/cnblogs_com/ikaros-521/2328032/o_230926154835_%E5%BE%AE%E4%BF%A1%E5%9B%BE%E7%89%87_20230926234750.jpg",
+            amount: "￥100"
+        },
         {
             name: "很养眼的丑男",
             avatar: "https://images.cnblogs.com/cnblogs_com/ikaros-521/2328032/o_230722162732_QQ%E5%9B%BE%E7%89%8720230723000544.jpg",

main.py

@@ -1673,6 +1673,15 @@ def play_audio_gui_create():
                 }
                 data_json.append(tmp_json)
 
+                tmp_json = {
+                    "label_text": "音频输出路径",
+                    "label_tip": "音频文件合成后存储的路径，支持相对路径或绝对路径",
+                    "data": play_audio_config["out_path"],
+                    "main_obj_name": "play_audio",
+                    "index": 1
+                }
+                data_json.append(tmp_json)
+
                 widgets = self.create_widgets_from_json(data_json)
 
                 # 动态添加widget到对应的gridLayout
@@ -3030,7 +3039,8 @@ def reorganize_copywriting_config_data(copywriting_config_data):
             config_data["database"] = reorganize_grid_data(database_data, database_keys_mapping)
 
             play_audio_keys_mapping = {
-                "enable": 0
+                "enable": 0,
+                "out_path": 1
             }
 
             # 重组play_audio数据并写回json

utils/audio.py

@@ -429,16 +429,11 @@ async def voice_change_and_put_to_queue(message, voice_tmp_path):
                 }
 
                 # 调用接口合成语音
-                data_json = self.my_tts.vits_fast_api(data)
+                voice_tmp_path = self.my_tts.vits_fast_api(data)
                 # logging.info(data_json)
 
-                if data_json is None:
-                    return
-
-                if "data" in data_json:
-                    voice_tmp_path = data_json["data"][1]["name"]
-                else:
-                    logging.error(f"vits-fast合成失败，请检查配置是否正确，接口返回：{data_json}")
+                if voice_tmp_path is None:
+                    logging.error(f"vits-fast合成失败，请检查配置是否正确")
                     return
 
                 logging.info(f"vits-fast合成成功，合成内容：【{message['content']}】，输出到={voice_tmp_path}")
@@ -449,12 +444,26 @@ async def voice_change_and_put_to_queue(message, voice_tmp_path):
                 return
         elif message["tts_type"] == "edge-tts":
             try:
-                voice_tmp_path = './out/' + self.common.get_bj_time(4) + '.mp3'
-                # 过滤" '字符
-                message["content"] = message["content"].replace('"', '').replace("'", '')
-                # 使用 Edge TTS 生成回复消息的语音文件
-                communicate = edge_tts.Communicate(text=message["content"], voice=message["data"]["voice"], rate=message["data"]["rate"], volume=message["data"]["volume"])
-                await communicate.save(voice_tmp_path)
+                data = {
+                    "content": message["content"],
+                    "voice": message["data"]["voice"],
+                    "rate": message["data"]["rate"],
+                    "volume": message["volume"]
+                }
+
+                # 调用接口合成语音
+                voice_tmp_path = await self.my_tts.edge_tts_api(data)
+
+                if voice_tmp_path is None:
+                    logging.error(f"edge-tts合成失败，请检查配置是否正确 或 网络问题")
+                    return
+
+                # voice_tmp_path = './out/' + self.common.get_bj_time(4) + '.mp3'
+                # # 过滤" '字符
+                # message["content"] = message["content"].replace('"', '').replace("'", '')
+                # # 使用 Edge TTS 生成回复消息的语音文件
+                # communicate = edge_tts.Communicate(text=message["content"], voice=message["data"]["voice"], rate=message["data"]["rate"], volume=message["data"]["volume"])
+                # await communicate.save(voice_tmp_path)
 
                 logging.info(f"edge-tts合成成功，合成内容：【{message['content']}】，输出到={voice_tmp_path}")
 
@@ -957,10 +966,7 @@ async def voice_change_and_put_to_queue(voice_tmp_path):
                         }
 
                         # 调用接口合成语音
-                        data_json = self.my_tts.vits_fast_api(data)
-                        # logging.info(data_json)
-
-                        voice_tmp_path = data_json["data"][1]["name"]
+                        voice_tmp_path = self.my_tts.vits_fast_api(data)
                         logging.info(f"vits-fast合成成功，合成内容：【{content}】，输出到={voice_tmp_path}")
 
                         await voice_change_and_put_to_queue(voice_tmp_path)

utils/audio_handle/my_tts.py

@@ -1,7 +1,9 @@
-import json, logging
+import json, logging, os
 import aiohttp, requests
 from urllib.parse import urlencode
 from gradio_client import Client
+import traceback
+import edge_tts
 
 from utils.common import Common
 from utils.logger import Configure_logger
@@ -16,6 +18,16 @@ def __init__(self, config_path):
         file_path = "./log/log-" + self.common.get_bj_time(1) + ".txt"
         Configure_logger(file_path)
 
+        try:
+            self.audio_out_path = self.config.get("play_audio", "out_path")
+
+            if not os.path.isabs(self.audio_out_path):
+                if not self.audio_out_path.startswith('./'):
+                    self.audio_out_path = './' + self.audio_out_path
+        except Exception as e:
+            logging.error(traceback.format_exc())
+            logging.error("请检查播放音频的音频输出路径配置！！！这将影响程序使用！")
+
 
     # 请求vits的api
     async def vits_api(self, data):
@@ -83,7 +95,8 @@ async def vits_api(self, data):
                 async with session.get(url) as response:
                     response = await response.read()
                     # print(response)
-                    voice_tmp_path = './out/vits_' + self.common.get_bj_time(4) + '.wav'
+                    file_name = 'vits_' + self.common.get_bj_time(4) + '.wav'
+                    voice_tmp_path = self.get_new_audio_path(file_name)
                     with open(voice_tmp_path, 'wb') as f:
                         f.write(response)
 
@@ -123,18 +136,35 @@ def vits_fast_api(self, data):
 
             result = response.content
             ret = json.loads(result)
-            return ret
-            # async with aiohttp.ClientSession() as session:
-            #     async with session.post(url=API_URL, json=data_json) as response:
-            #         result = await response.read()
-            #         # logging.info(result)
-            #         ret = json.loads(result)
-            # return ret
+
+            file_path = ret["data"][1]["name"]
+
+            new_file_path = self.common.move_file(file_path, self.audio_out_path, 'vits_fast_' + self.common.get_bj_time(4))
+
+            return new_file_path
         except Exception as e:
             logging.error(e)
             return None
 
 
+    # 请求Edge-TTS接口获取合成后的音频路径
+    async def edge_tts_api(self, data):
+        try:
+            file_name = 'edge_tts_' + self.common.get_bj_time(4) + '.wav'
+            voice_tmp_path = self.get_new_audio_path(file_name)
+            # voice_tmp_path = './out/' + self.common.get_bj_time(4) + '.mp3'
+            # 过滤" '字符
+            data["content"] = data["content"].replace('"', '').replace("'", '')
+            # 使用 Edge TTS 生成回复消息的语音文件
+            communicate = edge_tts.Communicate(text=data["content"], voice=data["voice"], rate=data["rate"], volume=data["volume"])
+            await communicate.save(voice_tmp_path)
+
+            return voice_tmp_path
+        except Exception as e:
+            logging.error(e)
+            return None
+
+
     # 请求bark-gui的api
     def bark_gui_api(self, data):
         try:
@@ -152,7 +182,9 @@ def bark_gui_api(self, data):
                 fn_index=3
             )
 
-            return result
+            new_file_path = self.common.move_file(result, self.audio_out_path, 'bark_gui_' + self.common.get_bj_time(4))
+
+            return new_file_path
         except Exception as e:
             logging.error(f'bark_gui请求失败: {e}')
             return None
@@ -171,7 +203,9 @@ def vall_e_x_api(self, data):
 				fn_index=5
             )
 
-            return result[1]
+            new_file_path = self.common.move_file(result[1], self.audio_out_path, 'vall_e_x_' + self.common.get_bj_time(4))
+
+            return new_file_path
         except Exception as e:
             logging.error(f'vall_e_x_api请求失败: {e}')
             return None
@@ -196,7 +230,11 @@ async def genshinvoice_top_api(self, text):
             async with aiohttp.ClientSession() as session:
                 async with session.get(url, params=params) as response:
                     response = await response.read()
-                    voice_tmp_path = './out/genshinvoice_top_' + self.common.get_bj_time(4) + '.wav'
+                    # voice_tmp_path = os.path.join(self.audio_out_path, 'genshinvoice_top_' + self.common.get_bj_time(4) + '.wav')
+                    file_name = 'genshinvoice_top_' + self.common.get_bj_time(4) + '.wav'
+
+                    voice_tmp_path = self.get_new_audio_path(file_name)
+
                     with open(voice_tmp_path, 'wb') as f:
                         f.write(response)
 
@@ -207,3 +245,18 @@ async def genshinvoice_top_api(self, text):
             logging.error(f'genshinvoice.top未知错误: {e}')
 
         return None
+
+
+    # 获取新的音频路径
+    def get_new_audio_path(self, file_name):
+        # 判断路径是否为绝对路径
+        if os.path.isabs(self.audio_out_path):
+            # 如果是绝对路径，直接使用
+            voice_tmp_path = os.path.join(self.audio_out_path, file_name)
+        else:
+            # 如果不是绝对路径，检查是否包含 ./，如果不包含，添加 ./，然后拼接路径
+            if not self.audio_out_path.startswith('./'):
+                self.audio_out_path = './' + self.audio_out_path
+            voice_tmp_path = os.path.normpath(os.path.join(self.audio_out_path, file_name))
+
+        return voice_tmp_path

utils/common.py

@@ -375,6 +375,17 @@ def split_path_and_filename(self, file_path):
 
     # 移动文件到指定路径 src dest
     def move_file(self, source_path, destination_path, rename=None, format="wav"):
+        """移动文件到指定路径
+
+        Args:
+            source_path (str): 文件路径含文件名
+            destination_path (_type_): 目标文件夹
+            rename (str, optional): 文件名. Defaults to None.
+            format (str, optional): 文件格式（实际上只是个假拓展名）. Defaults to "wav".
+
+        Returns:
+            str: 输出到的完整路径含文件名
+        """
         # if os.path.exists(destination_path):
         #     # 如果目标位置已存在同名文件，则先将其移动到回收站
         #     send2trash(destination_path)
@@ -399,6 +410,8 @@ def move_file(self, source_path, destination_path, rename=None, format="wav"):
         shutil.move(source_path, destination_path)
         print(f"文件移动成功：{source_path} -> {destination_path}")
 
+        return destination_path
+
 
     # 从文件路径中提取出带有扩展名的文件名
     # 返回：str

utils/my_handle.py

@@ -435,8 +435,7 @@ def local_qa_handle(self, data):
 
                     # 根据 弹幕日志类型进行各类日志写入
                     if My_handle.config.get("comment_log_type") == "问答":
-                        f.write(
-                            f"[{user_name} 提问]:{content}\n[AI回复{user_name}]:{resp_content_joined}\n" + tmp_content)
+                        f.write(f"[{user_name} 提问]:{content}\n[AI回复{user_name}]:{resp_content_joined}\n" + tmp_content)
                     elif My_handle.config.get("comment_log_type") == "问题":
                         f.write(f"[{user_name} 提问]:{content}\n" + tmp_content)
                     elif My_handle.config.get("comment_log_type") == "回答":