add whisper api

app.py

@@ -5,6 +5,10 @@ from sentence_transformers import SentenceTransformer
 import faiss
 import numpy as np
 import gradio as gr
+import openai
+
+# 設置OpenAI API密鑰
+openai.api_key = 'sk-zK6OrDxP5DvDdAQqnR_nEuUL3UrZf_4W7qvYj1uphjT3BlbkFJdmZAxlxUCFv92NnnMwSB15FhpmiDZSfG2QPueobSQA'
 
 def load_or_create_model_and_embeddings(model_name, data_file, output_dir):
     model_path = os.path.join(output_dir, 'saved_model')
@@ -63,6 +67,20 @@ def search_and_format(query):
         formatted_results += f"   情資: {result['text']}\n"
         formatted_results += f"   命名實體: {', '.join(result['entity_groups'])}\n\n"
     return formatted_results
+
+def transcribe_audio(audio):
+    try:
+        # 將音頻文件上傳到Whisper API
+        with open(audio, "rb") as audio_file:
+            transcript = openai.Audio.transcribe("whisper-1", audio_file)
+        return transcript.text
+    except Exception as e:
+        return f"轉錄時發生錯誤: {str(e)}"
+
+def audio_to_search(audio):
+    transcription = transcribe_audio(audio)
+    return search_and_format(transcription), transcription
+
 # 示例問題
 example_queries = [
     "Tell me about recent cyber attacks from Russia",
@@ -89,13 +107,14 @@ custom_css = """
 # 創建Gradio界面
 with gr.Blocks(css=custom_css) as iface:
     gr.Markdown("# AskCTI")
-    gr.Markdown("輸入查詢以搜索相關威脅情報，將顯示前3個最相關的結果，包括實體組。")
+    gr.Markdown("輸入查詢或使用語音輸入以搜索相關威脅情報，將顯示前3個最相關的結果，包括實體組。")
     
     with gr.Row(equal_height=True):
         with gr.Column(scale=1, min_width=300):
-            query_input = gr.Textbox(lines=3, label="")
-            submit_btn = gr.Button("查詢")
-            #clear_btn = gr.Button("清除")
+            query_input = gr.Textbox(lines=3, label="文字查詢")
+            with gr.Row():
+                submit_btn = gr.Button("查詢")
+                audio_input = gr.Audio(source="microphone", type="filepath", label="語音輸入")
             
             gr.Markdown("### 範例查詢")
             for i in range(0, len(example_queries), 2):
@@ -107,10 +126,11 @@ with gr.Blocks(css=custom_css) as iface:
                             )
         
         with gr.Column(scale=2):
-            output = gr.Textbox(lines=20, label="")
+            output = gr.Textbox(lines=20, label="搜索結果")
+            transcription_output = gr.Textbox(lines=3, label="語音轉錄結果")
 
     submit_btn.click(search_and_format, inputs=[query_input], outputs=[output])
-    #clear_btn.click(lambda: "", outputs=[query_input])
+    audio_input.change(audio_to_search, inputs=[audio_input], outputs=[output, transcription_output])
 
 # 啟動Gradio界面
 iface.launch()