mirror of
https://github.com/xinnan-tech/xiaozhi-esp32-server.git
synced 2026-07-21 22:53:56 +08:00
75 lines
2.6 KiB
Python
75 lines
2.6 KiB
Python
import time
|
|
import os
|
|
from typing import Optional, Tuple, List
|
|
from aip import AipSpeech
|
|
from core.providers.asr.base import ASRProviderBase
|
|
from config.logger import setup_logging
|
|
from core.providers.asr.dto.dto import InterfaceType
|
|
|
|
TAG = __name__
|
|
logger = setup_logging()
|
|
|
|
|
|
class ASRProvider(ASRProviderBase):
|
|
def __init__(self, config: dict, delete_audio_file: bool = True):
|
|
super().__init__()
|
|
self.interface_type = InterfaceType.NON_STREAM
|
|
self.app_id = config.get("app_id")
|
|
self.api_key = config.get("api_key")
|
|
self.secret_key = config.get("secret_key")
|
|
|
|
dev_pid = config.get("dev_pid", "1537")
|
|
self.dev_pid = int(dev_pid) if dev_pid else 1537
|
|
|
|
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)
|
|
|
|
async def speech_to_text(
|
|
self, opus_data: List[bytes], session_id: str, artifacts=None
|
|
) -> Tuple[Optional[str], Optional[str]]:
|
|
"""将语音数据转换为文本"""
|
|
if not opus_data:
|
|
logger.bind(tag=TAG).warning("音频数据为空！")
|
|
return None, 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, None
|
|
|
|
if artifacts is None:
|
|
return "", None
|
|
|
|
start_time = time.time()
|
|
# 识别本地文件
|
|
result = self.client.asr(
|
|
artifacts.pcm_bytes,
|
|
"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, artifacts.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, artifacts.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, None
|