response = OPEN_AI_CLIENT.audio.speech.create(

app.py

@@ -269,22 +269,29 @@ def paragraph_save(refine_paragraph):
 
 def paragraph_save_and_tts(paragraph_text):
     """
-    Saves the paragraph text and generates an audio file using Hugging Face's TTS.
+    Saves the paragraph text and generates an audio file using OpenAI's TTS.
     """
-    # Instantiate the text-to-speech pipeline
-    tts = pipeline("text-to-speech", model="facebook/fastspeech2-en-ljspeech")
+    try:
+        # Call OpenAI's TTS API to generate speech from text
+        response = OPEN_AI_CLIENT.audio.speech.create(
+            model="tts-1",
+            voice="alloy",
+            input=paragraph_text,
+        )
 
-    # Generate speech
-    speech = tts(paragraph_text)
+        # Define the path for the audio file
+        audio_path = "/mnt/data/generated_audio_openai.mp3"
 
-    # Define the path for the audio file
-    audio_path = "/mnt/data/generated_audio.wav"
+        # Save the audio stream to a file
+        response.stream_to_file(audio_path)
 
-    # Save the audio file
-    sf.write(audio_path, speech["array"], speech["sampling_rate"])
+        # Return the path to the audio file along with the text
+        return paragraph_text, audio_path
 
-    # Return the path to the audio file along with the text
-    return paragraph_text, audio_path
+    except Exception as e:
+        print(f"An error occurred while generating TTS: {e}")
+        # Handle the error appropriately (e.g., return an error message or a default audio path)
+        return paragraph_text, None
 
 with gr.Blocks() as demo:
     with gr.Row():

requirements.txt

@@ -1,4 +1,2 @@
 gradio
 openai>=1.0.0
-transformers
-soundfile