添加百度asr支持

main/xiaozhi-server/config.yaml

@@ -258,6 +258,19 @@ ASR:
     access_key_id: 你的阿里云账号access_key_id
     access_key_secret: 你的阿里云账号access_key_secret
     output_dir: tmp/
+  BaiduASR:
+    # 可以在这里申请百度语音技术的AppID、API Key、Secret Key
+    # https://console.bce.baidu.com/ai-engine/old/#/ai/speech/app/list
+    # 启用前需要先安装依赖包：
+    # pip install baidu-aip==4.16.13
+    # pip install chardet==5.2.0
+    type: baidu
+    app_id: 你的百度语音技术AppID
+    api_key: 你的百度语音技术APIKey
+    secret_key: 你的百度语音技术SecretKey
+    # 语言参数，1537为普通话，具体参考：https://ai.baidu.com/ai-doc/SPEECH/0lbxfnc9b
+    dev_pid: 1537
+    output_dir: tmp/
 
 VAD:
   SileroVAD:

main/xiaozhi-server/core/providers/asr/baidu.py

@@ -0,0 +1,103 @@
+import base64
+import hashlib
+import hmac
+import json
+import time
+from datetime import datetime, timezone
+import os
+import uuid
+from typing import Optional, Tuple, List
+import wave
+import opuslib_next
+
+from aip import AipSpeech
+from core.providers.asr.base import ASRProviderBase
+from config.logger import setup_logging
+
+TAG = __name__
+logger = setup_logging()
+
+class ASRProvider(ASRProviderBase):
+    def __init__(self, config: dict, delete_audio_file: bool = True):
+        self.app_id = config.get("app_id")
+        self.api_key = config.get("api_key")
+        self.secret_key = config.get("secret_key")
+        self.dev_pid = config.get("dev_pid")
+        self.output_dir = config.get("output_dir")
+        self.delete_audio_file = delete_audio_file
+
+        self.client = AipSpeech(str(self.app_id), self.api_key, self.secret_key)
+
+        # 确保输出目录存在
+        os.makedirs(self.output_dir, exist_ok=True)
+
+    def save_audio_to_file(self, pcm_data: List[bytes], session_id: str) -> str:
+        """PCM数据保存为WAV文件"""
+        module_name = __name__.split(".")[-1]
+        file_name = f"asr_{module_name}_{session_id}_{uuid.uuid4()}.wav"
+        file_path = os.path.join(self.output_dir, file_name)
+
+        with wave.open(file_path, "wb") as wf:
+            wf.setnchannels(1)
+            wf.setsampwidth(2)  # 2 bytes = 16-bit
+            wf.setframerate(16000)
+            wf.writeframes(b"".join(pcm_data))
+
+        return file_path
+
+    @staticmethod
+    def decode_opus(opus_data: List[bytes]) -> bytes:
+        """将Opus音频数据解码为PCM数据"""
+
+        decoder = opuslib_next.Decoder(16000, 1)  # 16kHz, 单声道
+        pcm_data = []
+
+        for opus_packet in opus_data:
+            try:
+                pcm_frame = decoder.decode(opus_packet, 960)  # 960 samples = 60ms
+                pcm_data.append(pcm_frame)
+            except opuslib_next.OpusError as e:
+                logger.bind(tag=TAG).error(f"Opus解码错误: {e}", exc_info=True)
+
+        return pcm_data
+
+    async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
+        """将语音数据转换为文本"""
+        if not opus_data:
+            logger.bind(tag=TAG).warn("音频数据为空！")
+            return None, None
+
+        file_path = None
+        try:
+            # 检查配置是否已设置
+            if not self.app_id or not self.api_key or not self.secret_key:
+                logger.bind(tag=TAG).error("百度语音识别配置未设置，无法进行识别")
+                return None, file_path
+
+            # 将Opus音频数据解码为PCM
+            pcm_data = self.decode_opus(opus_data)
+            combined_pcm_data = b"".join(pcm_data)
+
+            # 判断是否保存为WAV文件
+            if self.delete_audio_file:
+                pass
+            else:
+                self.save_audio_to_file(pcm_data, session_id)
+
+            start_time = time.time()
+            # 识别本地文件
+            result = self.client.asr(combined_pcm_data, 'pcm', 16000, {
+                'dev_pid': str(self.dev_pid),
+            })
+
+            if result and result["err_no"] == 0:
+                logger.bind(tag=TAG).debug(f"百度语音识别耗时: {time.time() - start_time:.3f}s | 结果: {result}")
+                result = result["result"][0]
+                return result, file_path
+            else:
+                raise Exception(f"百度语音识别失败，错误码: {result['err_no']}，错误信息: {result['err_msg']}")
+                return None, file_path
+
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"处理音频时发生错误！{e}", exc_info=True)
+            return None, file_path

main/xiaozhi-server/requirements.txt

@@ -25,4 +25,6 @@ sherpa_onnx==1.11.0
 mcp==1.4.1
 cnlunar==0.2.0
 PySocks==1.7.1
-dashscope==1.23.1
+dashscope==1.23.1
+baidu-aip==4.16.13
+chardet==5.2.0