feat: remove asr

app.py

@@ -4,7 +4,7 @@ Step-Audio-EditX - Audio Editing Demo using StepFun API
 import logging
 import gradio as gr
 
-from stepfun_api import get_api_token, process_audio, transcribe_audio
+from stepfun_api import get_api_token, process_audio  # transcribe_audio
 
 # Configure logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
@@ -170,53 +170,54 @@ class EditxTab:
         state["history_audio"] = []
         return [], state
 
-    def auto_transcribe_audio(self, audio_path, state):
-        """
-        自动转录音频文件，一次性返回最终结果
+    # def auto_transcribe_audio(self, audio_path, state):
+    #     """
+    #     自动转录音频文件，一次性返回最终结果
 
-        Args:
-            audio_path: 音频文件路径
-            state: 状态字典
+    #     Args:
+    #         audio_path: 音频文件路径
+    #         state: 状态字典
 
-        Returns:
-            转录的文本内容和更新后的状态
-        """
-        if not audio_path:
-            return "", state
+    #     Returns:
+    #         转录的文本内容和更新后的状态
+    #     """
+    #     if not audio_path:
+    #         return "", state
 
-        # 防止重复调用 - 简化逻辑
-        if state.get("last_audio_path") == audio_path:
-            self.logger.debug(f"⚠️ Skipping duplicate transcription request for: {audio_path}")
-            return state.get("last_transcribed_text", ""), state
+    #     # 防止重复调用 - 简化逻辑
+    #     if state.get("last_audio_path") == audio_path:
+    #         self.logger.debug(f"⚠️ Skipping duplicate transcription request for: {audio_path}")
+    #         return state.get("last_transcribed_text", ""), state
 
-        try:
-            # 更新音频路径
-            state["last_audio_path"] = audio_path
+    #     try:
+    #         # 更新音频路径
+    #         state["last_audio_path"] = audio_path
 
-            self.logger.info(f"🎙️ Starting auto transcription for: {audio_path}")
+    #         self.logger.info(f"🎙️ Starting auto transcription for: {audio_path}")
 
-            # 使用stepfun_api中的transcribe_audio函数，不使用streaming模式
-            transcribed_text = transcribe_audio(audio_path, streaming=False)
+    #         # 使用stepfun_api中的transcribe_audio函数，不使用streaming模式
+    #         transcribed_text = transcribe_audio(audio_path, streaming=False)
 
-            # 转录完成，缓存结果
-            state["last_transcribed_text"] = transcribed_text
+    #         # 转录完成，缓存结果
+    #         state["last_transcribed_text"] = transcribed_text
 
-            self.logger.info(f"✅ Auto transcription completed: {transcribed_text}")
-            return transcribed_text, state
+    #         self.logger.info(f"✅ Auto transcription completed: {transcribed_text}")
+    #         return transcribed_text, state
 
-        except Exception as e:
-            error_msg = f"[转录失败: {str(e)}]"
-            self.logger.error(f"❌ Auto transcription failed: {str(e)}")
-            state["last_transcribed_text"] = error_msg
-            return error_msg, state
+    #     except Exception as e:
+    #         error_msg = f"[转录失败: {str(e)}]"
+    #         self.logger.error(f"❌ Auto transcription failed: {str(e)}")
+    #         state["last_transcribed_text"] = error_msg
+    #         return error_msg, state
 
     def init_state(self):
         """Initialize conversation state"""
         return {
             "history_messages": [],
-            "history_audio": [],
-            "last_audio_path": None,  # 用于防重复调用
-            "last_transcribed_text": ""  # 缓存最后的转录结果
+            "history_audio": []
+            # # ASR相关状态（已禁用）
+            # "last_audio_path": None,  # 用于防重复调用
+            # "last_transcribed_text": ""  # 缓存最后的转录结果
         }
 
     def update_edit_info(self, category):
@@ -295,12 +296,12 @@ class EditxTab:
             outputs=self.edit_info,
         )
 
-        # 音频上传时自动转录
-        self.prompt_audio_input.change(
-            fn=self.auto_transcribe_audio,
-            inputs=[self.prompt_audio_input, state],
-            outputs=[self.prompt_text_input, state]
-        )
+        # # 音频上传时自动转录
+        # self.prompt_audio_input.change(
+        #     fn=self.auto_transcribe_audio,
+        #     inputs=[self.prompt_audio_input, state],
+        #     outputs=[self.prompt_text_input, state]
+        # )
 
 
 def create_demo():