接入Dify 聊天助手的API

.vscode/setting.json

@@ -0,0 +1,5 @@
+{
+    "ruff.args": [
+      "--ignore=F401,E402,F541"
+    ]
+  }

README.md

@@ -67,8 +67,6 @@ https://lug.ustc.edu.cn/wiki/doc/smart-questions
 
 ### 🌏 合作伙伴
 
-青椒云：[qingjiaocloud.com](https://account.qingjiaocloud.com/signin?inviteCode=4Q92ROBQ)  ———— AIGC 高性能云桌面  
-
 睿声AI：[reecho.ai](https://www.reecho.ai/)  ———— 瞬时创建 超级拟真的 人工智能克隆语音  
 
 AIHubMix: [aihubmix.com](https://aihubmix.com/register?aff=1BMI)  ———— OpenAI,Google,通义千问等大语言模型API代理站  

config.json

@@ -530,6 +530,12 @@
     "mode": "chat",
     "workspace_slug": "test"
   },
+  "dify": {
+    "api_ip_port": "http://172.26.189.21",
+    "type": "聊天助手",
+    "api_key": "app-64xu0vQjP2kxN4DKR8Ch7ZGY",
+    "history_enable": true
+  },
   "llm_tpu": {
     "api_ip_port": "http://127.0.0.1:8003",
     "max_length": 1,
@@ -1838,6 +1844,7 @@
         "koboldcpp": true,
         "anythingllm": true,
         "gpt4free": true,
+        "dify": true,
         "llm_tpu": true,
         "custom_llm": true
       },

config.json.bak

@@ -530,6 +530,12 @@
     "mode": "chat",
     "workspace_slug": "test"
   },
+  "dify": {
+    "api_ip_port": "http://172.26.189.21",
+    "type": "聊天助手",
+    "api_key": "app-64xu0vQjP2kxN4DKR8Ch7ZGY",
+    "history_enable": true
+  },
   "llm_tpu": {
     "api_ip_port": "http://127.0.0.1:8003",
     "max_length": 1,
@@ -1838,6 +1844,7 @@
         "koboldcpp": true,
         "anythingllm": true,
         "gpt4free": true,
+        "dify": true,
         "llm_tpu": true,
         "custom_llm": true
       },

tests/test_dify/api.py

@@ -0,0 +1,86 @@
+import json, logging
+import re, requests
+import traceback
+from urllib.parse import urljoin
+import sys
+sys.path.insert(1, "../../utils")
+#from utils.common import Common
+from loguru import logger
+
+
+class Dify:
+    def __init__(self, data):
+        #self.common = Common()
+        self.config_data = data
+
+        self.conversation_id = ""
+
+        logger.debug(self.config_data)
+
+
+    def get_resp(self, data):
+        """请求对应接口，获取返回值
+
+        Args:
+            data (dict): 含有提问的json数据
+
+        Returns:
+            str: 返回的文本回答
+        """
+        try:
+            resp_content = None
+
+            if self.config_data["type"] == "聊天助手":
+                API_URL = urljoin(self.config_data["api_ip_port"], '/v1/chat-messages')
+
+                data_json = {
+                    "inputs": {},
+                    "query": data["prompt"],
+                    # 阻塞模式
+                    "response_mode": "blocking",
+                    # 会话 ID，需要基于之前的聊天记录继续对话，必须传之前消息的 conversation_id。
+                    "conversation_id": self.conversation_id,
+                    # 用户名是否区分 视情况而定，暂时为了稳定性统一
+                    "user": "test"
+                }
+                headers = {
+                    'Content-Type': 'application/json',
+                    'Authorization': f'Bearer {self.config_data["api_key"]}'
+                }
+
+                response = requests.request("POST", API_URL, headers=headers, json=data_json)
+                resp_json = json.loads(response.content)
+
+                logger.debug(f"resp_json={resp_json}")
+
+                if "answer" in resp_json:
+                    resp_content = resp_json["answer"]
+
+                    # 是否记录历史
+                    if self.config_data["history_enable"]:
+                        self.conversation_id = resp_json["conversation_id"]
+                else:
+                    logger.error(f"获取LLM返回失败。{resp_json}")
+                    return None
+
+                return resp_content
+
+        except Exception as e:
+            logger.error(traceback.format_exc())
+
+        return None
+
+if __name__ == '__main__':
+
+
+    data = {
+        "api_ip_port": "http://172.26.189.21/v1",
+        "type": "聊天助手",
+        "api_key": "app-64xu0vQjP2kxN4DKR8Ch7ZGY",
+        "history_enable": True
+    }
+
+    # 实例化并调用
+    dify = Dify(data)
+    logger.info(dify.get_resp({"prompt": "你可以扮演猫娘吗，每句话后面加个喵"}))
+    logger.info(dify.get_resp({"prompt": "早上好"}))

tests/test_gpt_sovits/api_v3.py

@@ -0,0 +1,99 @@
+from typing import Iterator
+import time
+import requests
+import numpy as np
+import resampy
+import pyaudio
+
+class VitsTTS:
+    audio = None
+    stream = None
+    sample_rate = 16000  # Set the desired sample rate for playback
+
+    def __init__(self, config_json):
+        self.config_json = config_json
+
+        # Initialize PyAudio and stream if not already done
+        if VitsTTS.audio is None:
+            VitsTTS.audio = pyaudio.PyAudio()
+        if VitsTTS.stream is None or not VitsTTS.stream.is_active():
+            VitsTTS.stream = VitsTTS.audio.open(
+                format=pyaudio.paFloat32,
+                channels=1,
+                rate=VitsTTS.sample_rate,
+                output=True
+            )
+
+    def txt_to_audio(self, msg): 
+        self.stream_tts(
+            self.gpt_sovits(
+                msg,
+                self.config_json["ref_file"],  
+                self.config_json["ref_text"],
+                "zh",  # Language (can be "en" or other supported languages)
+                self.config_json["server_url"]
+            )
+        )
+
+    def gpt_sovits(self, text, reffile, reftext, language, server_url) -> Iterator[bytes]:
+        start = time.perf_counter()
+        req = {
+            'text': text,
+            'text_lang': language,
+            'ref_audio_path': reffile,
+            'prompt_text': reftext,
+            'prompt_lang': language,
+            'media_type': 'raw',
+            'streaming_mode': True,
+        }
+
+        res = requests.post(
+            f"{server_url}/tts",
+            json=req,
+            stream=True,
+        )
+        end = time.perf_counter()
+        print(f"gpt_sovits Time to make POST: {end-start}s")
+
+        if res.status_code != 200:
+            print("Error:", res.text)
+            return
+
+        first = True
+        for chunk in res.iter_content(chunk_size=32000):  # 32K*20ms*2
+            if first:
+                end = time.perf_counter()
+                print(f"gpt_sovits Time to first chunk: {end-start}s")
+                first = False
+            if chunk:
+                yield chunk
+
+        print("gpt_sovits response.elapsed:", res.elapsed)
+
+    def stream_tts(self, audio_stream):
+        for chunk in audio_stream:
+            if chunk is not None and len(chunk) > 0:
+                stream = np.frombuffer(chunk, dtype=np.int16).astype(np.float32) / 32767
+                stream = resampy.resample(x=stream, sr_orig=32000, sr_new=VitsTTS.sample_rate)
+                VitsTTS.stream.write(stream.tobytes())
+
+    @classmethod
+    def close_audio(cls):
+        if cls.stream is not None:
+            cls.stream.stop_stream()
+            cls.stream.close()
+            cls.audio.terminate()
+            cls.stream = None
+            cls.audio = None
+
+
+if __name__ == "__main__":
+    config_json = {
+        "server_url": "http://127.0.0.1:9880",
+        "ref_file": "E:\\GitHub_pro\\AI-Vtuber\\out\\edge_tts_3.mp3",
+        "ref_text": "就送个人气票，看不起谁呢",
+    }
+    vits_tts = VitsTTS(config_json)
+    vits_tts.txt_to_audio("你好，我是AI")
+    vits_tts.txt_to_audio("我的声音如何")
+    vits_tts.txt_to_audio("床前明月光，疑是地上霜。举头望明月，低头思故乡。")

utils/gpt_model/dify.py

@@ -0,0 +1,80 @@
+import json
+import requests
+import traceback
+from urllib.parse import urljoin
+from loguru import logger
+
+
+class Dify:
+    def __init__(self, data: dict):
+        self.config_data = data
+
+        self.conversation_id = ""
+
+        # logger.debug(self.config_data)
+
+
+    def get_resp(self, data: dict):
+        """请求对应接口，获取返回值
+
+        Args:
+            data (dict): 含有提问的json数据
+
+        Returns:
+            str: 返回的文本回答
+        """
+        try:
+            resp_content = None
+
+            if self.config_data["type"] == "聊天助手":
+                API_URL = urljoin(self.config_data["api_ip_port"], '/v1/chat-messages')
+
+                data_json = {
+                    "inputs": {},
+                    "query": data["prompt"],
+                    # 阻塞模式
+                    "response_mode": "blocking",
+                    # 会话 ID，需要基于之前的聊天记录继续对话，必须传之前消息的 conversation_id。
+                    "conversation_id": self.conversation_id,
+                    # 用户名是否区分 视情况而定，暂时为了稳定性统一
+                    "user": "test"
+                }
+                headers = {
+                    'Content-Type': 'application/json',
+                    'Authorization': f'Bearer {self.config_data["api_key"]}'
+                }
+
+                response = requests.request("POST", API_URL, headers=headers, json=data_json)
+                resp_json = json.loads(response.content)
+
+                logger.debug(f"resp_json={resp_json}")
+
+                if "answer" in resp_json:
+                    resp_content = resp_json["answer"]
+
+                    # 是否记录历史
+                    if self.config_data["history_enable"]:
+                        self.conversation_id = resp_json["conversation_id"]
+                else:
+                    logger.error(f"获取LLM返回失败。{resp_json}")
+                    return None
+
+                return resp_content
+
+        except Exception as e:
+            logger.error(traceback.format_exc())
+
+        return None
+
+if __name__ == '__main__':
+    data = {
+        "api_ip_port": "http://172.26.189.21/v1",
+        "type": "聊天助手",
+        "api_key": "app-64xu0vQjP2kxN4DKR8Ch7ZGY",
+        "history_enable": True
+    }
+
+    # 实例化并调用
+    dify = Dify(data)
+    logger.info(dify.get_resp({"prompt": "你可以扮演猫娘吗，每句话后面加个喵"}))
+    logger.info(dify.get_resp({"prompt": "早上好"}))

utils/gpt_model/gpt.py

@@ -32,6 +32,7 @@
 from utils.gpt_model.gpt4free import GPT4Free
 from utils.gpt_model.custom_llm import Custom_LLM
 from utils.gpt_model.llm_tpu import LLM_TPU
+from utils.gpt_model.dify import Dify
 
 class GPT_Model:
     openai = None
@@ -60,6 +61,7 @@ def set_model_config(self, model_name, config):
             "gpt4free": GPT4Free,
             "custom_llm": Custom_LLM,
             "llm_tpu": LLM_TPU,
+            "dify": Dify,
         }
 
         if model_name == "openai":

utils/my_handle.py

@@ -166,13 +166,14 @@ def __init__(self, config_path):
             self.gpt4free = None
             self.custom_llm = None
             self.llm_tpu = None
+            self.dify = None
 
             self.image_recognition_model = None
 
             self.chat_type_list = ["chatgpt", "claude", "claude2", "chatglm", "qwen", "chat_with_file", "text_generation_webui", \
                     "sparkdesk", "langchain_chatglm", "langchain_chatchat", "zhipu", "bard", "yiyan", "tongyi", \
                     "tongyixingchen", "my_qianfan", "my_wenxinworkshop", "gemini", "qanything", "koboldcpp", "anythingllm", "gpt4free", \
-                    "custom_llm", "llm_tpu"]
+                    "custom_llm", "llm_tpu", "dify"]
 
             # 配置加载
             self.config_load()
@@ -1530,6 +1531,7 @@ def llm_handle(self, chat_type, data, type="chat", webui_show=True):
                     "gpt4free": lambda: self.gpt4free.get_resp({"prompt": data["content"]}),
                     "custom_llm": lambda: self.custom_llm.get_resp({"prompt": data["content"]}),
                     "llm_tpu": lambda: self.llm_tpu.get_resp({"prompt": data["content"]}),
+                    "dify": lambda: self.dify.get_resp({"prompt": data["content"]}),
                     "reread": lambda: data["content"]
                 }
             elif type == "vision":