update:添加vosk

2026-07-26 17:13:54 +08:00 · 2025-08-27 14:39:05 +08:00
parent b8803f92e7
commit cc7ba0872c
4 changed files with 152 additions and 0 deletions
@@ -0,0 +1,114 @@
+import os
+import json
+import time
+from typing import Optional, Tuple, List
+from .base import ASRProviderBase
+from config.logger import setup_logging
+from core.providers.asr.dto.dto import InterfaceType
+import vosk
+
+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.model_path = config.get("model_path", "models/vosk/vosk-model-small-cn-0.22")
+        self.output_dir = config.get("output_dir", "tmp/")
+        self.delete_audio_file = delete_audio_file
+        
+        # 初始化VOSK模型
+        self.model = None
+        self.recognizer = None
+        self._load_model()
+        
+        # 确保输出目录存在
+        os.makedirs(self.output_dir, exist_ok=True)
+
+    def _load_model(self):
+        """加载VOSK模型"""
+        try:
+            if not os.path.exists(self.model_path):
+                raise FileNotFoundError(f"VOSK模型路径不存在: {self.model_path}")
+                
+            logger.bind(tag=TAG).info(f"正在加载VOSK模型: {self.model_path}")
+            self.model = vosk.Model(self.model_path)
+
+            # 初始化VOSK识别器（采样率必须为16kHz）
+            self.recognizer = vosk.KaldiRecognizer(self.model, 16000)
+
+            logger.bind(tag=TAG).info("VOSK模型加载成功")
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"加载VOSK模型失败: {e}")
+            raise
+
+    async def speech_to_text(
+        self, audio_data: List[bytes], session_id: str, audio_format: str = "opus"
+    ) -> Tuple[Optional[str], Optional[str]]:
+        """将语音数据转换为文本"""
+        file_path = None
+        try:
+            # 检查模型是否加载成功
+            if not self.model:
+                logger.bind(tag=TAG).error("VOSK模型未加载，无法进行识别")
+                return "", None
+
+            # 解码音频（如果原始格式是Opus）
+            if audio_format == "pcm":
+                pcm_data = audio_data
+            else:
+                pcm_data = self.decode_opus(audio_data)
+                
+            if not pcm_data:
+                logger.bind(tag=TAG).warning("解码后的PCM数据为空，无法进行识别")
+                return "", None
+                
+            # 合并PCM数据
+            combined_pcm_data = b"".join(pcm_data)
+            if len(combined_pcm_data) == 0:
+                logger.bind(tag=TAG).warning("合并后的PCM数据为空")
+                return "", None
+
+            # 判断是否保存为WAV文件
+            if not self.delete_audio_file:
+                file_path = self.save_audio_to_file(pcm_data, session_id)
+
+            start_time = time.time()
+            
+            
+            # 进行识别（VOSK推荐每次送入2000字节的数据）
+            chunk_size = 2000
+            text_result = ""
+            
+            for i in range(0, len(combined_pcm_data), chunk_size):
+                chunk = combined_pcm_data[i:i+chunk_size]
+                if self.recognizer.AcceptWaveform(chunk):
+                    result = json.loads(self.recognizer.Result())
+                    text = result.get('text', '')
+                    if text:
+                        text_result += text + " "
+            
+            # 获取最终结果
+            final_result = json.loads(self.recognizer.FinalResult())
+            final_text = final_result.get('text', '')
+            if final_text:
+                text_result += final_text
+            
+            logger.bind(tag=TAG).debug(
+                f"VOSK语音识别耗时: {time.time() - start_time:.3f}s | 结果: {text_result.strip()}"
+            )
+            
+            return text_result.strip(), file_path
+            
+        except Exception as e:
+            logger.bind(tag=TAG).error(f"VOSK语音识别失败: {e}")
+            return "", None
+        finally:
+            # 文件清理逻辑
+            if self.delete_audio_file and file_path and os.path.exists(file_path):
+                try:
+                    os.remove(file_path)
+                    logger.bind(tag=TAG).debug(f"已删除临时音频文件: {file_path}")
+                except Exception as e:
+                    logger.bind(tag=TAG).error(f"文件删除失败: {file_path} | 错误: {e}")