Update openai_tts_tool.py

openai_tts_tool.py

@@ -1,42 +1,29 @@
 import openai
-import os
 
-# Ensure you load your API key from the environment or a .env file
-openai_api_key = os.getenv("OPENAI_API_KEY")
+def generate_audio_and_text(api_key, input_text, model_name, voice_type, voice_speed, language, output_option, summary_length, additional_prompt):
+    # Set API key dynamically
+    openai.api_key = api_key
+    client = openai.OpenAI(api_key=api_key)
 
-# Ensure the OpenAI client is set up correctly
-client = openai.Client(api_key=openai_api_key)
+    # Assuming text-to-speech and summarization logic goes here
+    if output_option in ["summary_text", "both"]:
+        text_summary = f"Generated summary for: {input_text[:100]}..."  # Replace with real summary generation logic
 
-def generate_audio_and_text(api_key, text, model_name, voice_type, voice_speed, language, output_option, summary_length, additional_prompt):
-    """
-    Generate both text and audio based on the input parameters.
-    - text: The text input for processing.
-    - output_option: Either 'Text', 'Audio', or 'Both' for output.
-    """
+    if output_option in ["audio", "both"]:
+        response = client.audio.speech.create(
+            text=input_text,
+            model=model_name,
+            voice=voice_type,
+            language=language,
+            speed=voice_speed
+        )
+        audio_output = response['audio_file']  # Placeholder for the actual audio file output
 
-    # Placeholder for returning text (summary or full)
-    generated_text = text
-    
-    # Generate audio only if audio is requested
-    audio_path = None
+    if output_option == "summary_text":
+        return text_summary, None
+    elif output_option == "audio":
+        return None, audio_output
+    elif output_option == "both":
+        return text_summary, audio_output
 
-    if output_option in ['Audio', 'Both']:
-        audio_path = "output_audio.wav"  # Define the output path for the audio
-        try:
-            response = client.audio.speech.create(
-                model="tts-1-hd",  # Use the updated model for TTS
-                voice=voice_type,
-                input=text
-            )
-            response.stream_to_file(audio_path)
-        except Exception as e:
-            return f"Error generating audio: {str(e)}", None, None
-
-    # If 'Text' or 'Both' is selected, return the text as well
-    if output_option in ['Text', 'Both']:
-        if summary_length:  # If the user specifies a summary, we would summarize the text here
-            generated_text = f"Summarized text: {text[:summary_length]}..."  # Placeholder for summary logic
-        else:
-            generated_text = text
-
-    return generated_text, audio_path
+    return None, None