xiaozhi-sdk 0.0.8__tar.gz → 0.0.10__tar.gz

{xiaozhi_sdk-0.0.8/xiaozhi_sdk.egg-info → xiaozhi_sdk-0.0.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: xiaozhi-sdk
-Version: 0.0.8
+Version: 0.0.10
 Summary: 一个用于连接和控制小智智能设备的Python SDK，支持实时音频通信、MCP工具集成和设备管理功能。
 Author-email: dairoot <623815825@qq.com>
 License-Expression: MIT

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/tests/test_xiaozhi.py

@@ -69,9 +69,6 @@ async def message_handler_callback(message):
 
 MAC_ADDR = "00:22:44:66:88:00"
 
-ota_url = "http://localhost:3080/api/ota"
-URL = "ws://120.79.156.134:8380"
-
 ota_url = None
 URL = None
 
@@ -86,21 +83,21 @@ async def test_main():
     await xiaozhi.init_connection(MAC_ADDR)
 
     # # say hellow
-    # for pcm in read_audio_file("./file/audio/say_hello.wav"):
-    #     await xiaozhi.send_audio(pcm)
-    # await xiaozhi.send_silence_audio()
-    # await assistant_audio_play(xiaozhi.output_audio_queue)
-
-    # # say take photo
-    # for pcm in read_audio_file("./file/audio/take_photo.wav"):
-    #     await xiaozhi.send_audio(pcm)
-    # await xiaozhi.send_silence_audio()
-    # await assistant_audio_play(xiaozhi.output_audio_queue, 5)
+    for pcm in read_audio_file("./file/audio/say_hello.wav"):
+        await xiaozhi.send_audio(pcm)
+    await xiaozhi.send_silence_audio()
+    await assistant_audio_play(xiaozhi.output_audio_queue)
 
-    # play music
-    for pcm in read_audio_file("./file/audio/play_music.wav"):
+    # say take photo
+    for pcm in read_audio_file("./file/audio/take_photo.wav"):
         await xiaozhi.send_audio(pcm)
     await xiaozhi.send_silence_audio()
-    await assistant_audio_play(xiaozhi.output_audio_queue, 500)
+    await assistant_audio_play(xiaozhi.output_audio_queue, 5)
+
+    # play music
+    # for pcm in read_audio_file("./file/audio/play_music.wav"):
+    #     await xiaozhi.send_audio(pcm)
+    # await xiaozhi.send_silence_audio()
+    # await assistant_audio_play(xiaozhi.output_audio_queue, 500)
 
     await xiaozhi.close()

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/__init__.py

@@ -1,3 +1,3 @@
-__version__ = "0.0.8"
+__version__ = "0.0.10"
 
 from xiaozhi_sdk.core import XiaoZhiWebsocket  # noqa

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/cli.py

@@ -12,6 +12,34 @@ import sounddevice as sd
 from xiaozhi_sdk import XiaoZhiWebsocket
 from xiaozhi_sdk.config import INPUT_SERVER_AUDIO_SAMPLE_RATE
 
+# 定义自定义日志级别
+INFO1 = 21
+INFO2 = 22
+INFO3 = 23
+
+# 添加自定义日志级别到logging模块
+logging.addLevelName(INFO1, "INFO1")
+logging.addLevelName(INFO2, "INFO2")
+logging.addLevelName(INFO3, "INFO3")
+
+# 为logger添加自定义方法
+def info1(self, message, *args, **kwargs):
+    if self.isEnabledFor(INFO1):
+        self._log(INFO1, message, args, **kwargs)
+
+def info2(self, message, *args, **kwargs):
+    if self.isEnabledFor(INFO2):
+        self._log(INFO2, message, args, **kwargs)
+
+def info3(self, message, *args, **kwargs):
+    if self.isEnabledFor(INFO3):
+        self._log(INFO3, message, args, **kwargs)
+
+# 将自定义方法添加到Logger类
+logging.Logger.info1 = info1
+logging.Logger.info2 = info2
+logging.Logger.info3 = info3
+
 # 配置彩色logging
 handler = colorlog.StreamHandler()
 handler.setFormatter(
@@ -19,8 +47,11 @@ handler.setFormatter(
         "%(log_color)s%(asctime)s - %(name)s - %(levelname)s - %(message)s",
         datefmt="%Y-%m-%d %H:%M:%S",
         log_colors={
-            "DEBUG": "green",
-            "INFO": "white",
+            "DEBUG": "white",
+            "INFO": "green",
+            "INFO1": "green",
+            "INFO2": "cyan",
+            "INFO3": "blue",
             "WARNING": "yellow",
             "ERROR": "red",
             "CRITICAL": "red,bg_white",
@@ -41,17 +72,27 @@ is_end = False
 async def handle_message(message):
     """处理接收到的消息"""
     global is_end
-    logger.info("message received: %s", message)
+    if message["type"] == "stt":
+        logger.info1("message received: %s", message)
+    elif message["type"] == "tts":
+        logger.info2("message received: %s", message)
+    elif message["type"] == "llm":
+        logger.info3("message received: %s", message)
+    else:
+        logger.info("message received: %s", message)
+
     if message["type"] == "websocket" and message["state"] == "close":
         is_end = True
 
 
-async def play_assistant_audio(audio_queue: deque[bytes]):
+async def play_assistant_audio(audio_queue: deque[bytes], enable_audio):
     """播放音频流"""
     global is_playing_audio
 
-    stream = sd.OutputStream(samplerate=INPUT_SERVER_AUDIO_SAMPLE_RATE, channels=1, dtype=np.int16)
-    stream.start()
+    stream = None
+    if enable_audio:
+        stream = sd.OutputStream(samplerate=INPUT_SERVER_AUDIO_SAMPLE_RATE, channels=1, dtype=np.int16)
+        stream.start()
     last_audio_time = None
 
     while True:
@@ -66,7 +107,8 @@ async def play_assistant_audio(audio_queue: deque[bytes]):
 
         is_playing_audio = True
         pcm_data = audio_queue.popleft()
-        stream.write(pcm_data)
+        if stream:
+            stream.write(pcm_data)
         last_audio_time = time.time()
 
 
@@ -83,7 +125,7 @@ class XiaoZhiClient:
         self.ota_url = ota_url
         self.mac_address = ""
 
-    async def start(self, mac_address: str, serial_number: str = "", license_key: str = ""):
+    async def start(self, mac_address: str, serial_number: str, license_key: str, enable_audio):
         """启动客户端连接"""
         self.mac_address = mac_address
         self.xiaozhi = XiaoZhiWebsocket(handle_message, url=self.url, ota_url=self.ota_url, send_wake=True)
@@ -92,7 +134,7 @@ class XiaoZhiClient:
             self.mac_address, aec=False, serial_number=serial_number, license_key=license_key
         )
 
-        asyncio.create_task(play_assistant_audio(self.xiaozhi.output_audio_queue))
+        asyncio.create_task(play_assistant_audio(self.xiaozhi.output_audio_queue, enable_audio))
 
     def audio_callback(self, indata, frames, time, status):
         """音频输入回调函数"""
@@ -115,11 +157,11 @@ class XiaoZhiClient:
                 await self.xiaozhi.send_audio(pcm_data)
 
 
-async def run_client(mac_address: str, url: str, ota_url: str, serial_number: str, license_key: str):
+async def run_client(mac_address: str, url: str, ota_url: str, serial_number: str, license_key: str, enable_audio: bool):
     """运行客户端的异步函数"""
     logger.debug("Recording... Press Ctrl+C to stop.")
     client = XiaoZhiClient(url, ota_url)
-    await client.start(mac_address, serial_number, license_key)
+    await client.start(mac_address, serial_number, license_key, enable_audio)
 
     with sd.InputStream(callback=client.audio_callback, channels=1, samplerate=16000, blocksize=960):
         await client.process_audio_input()
@@ -131,9 +173,10 @@ async def run_client(mac_address: str, url: str, ota_url: str, serial_number: st
 @click.option("--ota_url", help="OTA地址")
 @click.option("--serial_number", default="", help="设备的序列号")
 @click.option("--license_key", default="", help="设备的授权密钥")
-def main(mac_address: str, url: str, ota_url: str, serial_number: str, license_key: str):
+@click.option("--enable_audio", default=True, help="是否开启音频播放")
+def main(mac_address: str, url: str, ota_url: str, serial_number: str, license_key: str, enable_audio: bool):
     """小智SDK客户端
 
     MAC_ADDRESS: 设备的MAC地址 (格式: XX:XX:XX:XX:XX:XX)
     """
-    asyncio.run(run_client(mac_address, url, ota_url, serial_number, license_key))
+    asyncio.run(run_client(mac_address, url, ota_url, serial_number, license_key, enable_audio))

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/iot.py

@@ -12,7 +12,7 @@ from xiaozhi_sdk.config import OTA_URL
 # 常量定义
 BOARD_TYPE = "xiaozhi-sdk-box"
 USER_AGENT = "xiaozhi-sdk/{}".format(__version__)
-BOARD_NAME = "xiaozhi-sdk-{}".format(__version__)
+BOARD_NAME = "xiaozhi-sdk"
 
 logger = logging.getLogger("xiaozhi_sdk")
 
@@ -60,6 +60,9 @@ class OtaDevice:
 
         async with aiohttp.ClientSession() as session:
             async with session.post(self.ota_url + "/", headers=headers, data=json.dumps(payload)) as response:
+                if response.status != 200:
+                    err_text = await response.text()
+                    raise Exception(err_text)
                 response.raise_for_status()
                 return await response.json()
 

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/mcp.py

@@ -43,7 +43,7 @@ class McpTool(object):
         try:
             response = requests.post(self.explain_url, files=files, data=payload, headers=headers, timeout=5)
             res_json = response.json()
-        except Exception as e:
+        except Exception:
             return "网络异常", True
         if res_json.get("error"):
             return res_json, True
@@ -72,10 +72,11 @@ class McpTool(object):
                 else:
                     tool_res, is_error = {"message": "正在为你播放: {}".format(arguments["music_name"])}, False
                     data = {
-                        "type": "music", "state": "start",
+                        "type": "music",
+                        "state": "start",
                         "url": music_info["url"],
                         "text": arguments["music_name"],
-                        "source": "sdk.mcp_music_tool"
+                        "source": "sdk.mcp_music_tool",
                     }
                     await self.message_handler_callback(data)
 

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/utils/mcp_data.py

@@ -1,4 +1,4 @@
-from typing import Any, Dict, List
+from typing import Any, Dict
 
 mcp_initialize_payload: Dict[str, Any] = {
     "jsonrpc": "2.0",
@@ -23,7 +23,10 @@ mcp_tool_conf: Dict[str, Dict[str, Any]] = {
         "description": "Play music using music IDs. IMPORTANT: You must call `search_custom_music` first to get the music IDs before using this tool. Use this tool after getting music IDs from search results. Args:\n  `id_list`: The id list of the music to play (obtained from search_custom_music results). The list must contain more than 2 music IDs, and the system will randomly select one to play.\n  `music_name`: The name of the music (obtained from search_custom_music results)",
         "inputSchema": {
             "type": "object",
-            "properties": {"music_name": {"type": "string"}, "id_list": {"type": "array", "items": {"type": "string"}, "minItems": 3}},
+            "properties": {
+                "music_name": {"type": "string"},
+                "id_list": {"type": "array", "items": {"type": "string"}, "minItems": 3},
+            },
             "required": ["music_name", "id_list"],
         },
     },

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10}/xiaozhi_sdk/utils/mcp_tool.py

@@ -83,7 +83,7 @@ async def async_mcp_play_music(data) -> tuple[list, bool]:
 
     chunk_size = 960 * 2
     for i in range(0, len(pcm_data), chunk_size):
-        chunk = pcm_data[i: i + chunk_size]
+        chunk = pcm_data[i : i + chunk_size]
 
         if chunk:  # 确保不添加空块
             chunk = np.frombuffer(chunk, dtype=np.int16)

{xiaozhi_sdk-0.0.8 → xiaozhi_sdk-0.0.10/xiaozhi_sdk.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: xiaozhi-sdk
-Version: 0.0.8
+Version: 0.0.10
 Summary: 一个用于连接和控制小智智能设备的Python SDK，支持实时音频通信、MCP工具集成和设备管理功能。
 Author-email: dairoot <623815825@qq.com>
 License-Expression: MIT