Upload folder using huggingface_hub

app.py

@@ -16,10 +16,58 @@ def tts(
     """
     Convert input text to speech using OpenAI's Text-to-Speech API.
 
-    (Function definition remains the same)
+    Parameters:
+        input_text (str): The text to be converted to speech.
+        model (str): The model to use for synthesis (e.g., 'tts-1', 'tts-1-hd').
+        voice (str): The voice profile to use (e.g., 'alloy', 'echo', 'fable', etc.).
+        api_key (str): OpenAI API key.
+        response_format (str): The audio format of the output file (default is 'mp3').
+        speed (float): The speed of the synthesized speech.
+
+    Returns:
+        str: File path to the generated audio file.
+
+    Raises:
+        gr.Error: If input parameters are invalid or API call fails.
     """
-    # (Function body remains the same)
-    # ...
+    if not api_key.strip():
+        raise gr.Error(
+            "API key is required. Get an API key at: https://platform.openai.com/account/api-keys"
+        )
+
+    if not input_text.strip():
+        raise gr.Error("Input text cannot be empty.")
+
+    openai.api_key = api_key
+
+    try:
+        response = openai.Audio.create(
+            text=input_text,
+            voice=voice,
+            model=model,
+            response_format=response_format,
+            speed=speed,
+        )
+    except openai.OpenAIError as e:
+        # Catch-all for OpenAI exceptions
+        raise gr.Error(f"An OpenAI error occurred: {e}")
+    except Exception as e:
+        # Catch any other exceptions
+        raise gr.Error(f"An unexpected error occurred: {e}")
+
+    if not hasattr(response, "audio"):
+        raise gr.Error(
+            "Invalid response from OpenAI API. The response does not contain audio content."
+        )
+
+    # Save the audio content to a temporary file
+    audio_content = response.audio
+    file_extension = f".{response_format}"
+    with tempfile.NamedTemporaryFile(suffix=file_extension, delete=False) as temp_file:
+        temp_file.write(audio_content)
+        temp_file_path = temp_file.name
+
+    return temp_file_path
 
 def main():
     """
@@ -62,7 +110,8 @@ def main():
         value=VOICE_PREVIEW_FILES['echo'],
         visible=True,
         show_download_button=False,
-        show_share_button=False
+        show_share_button=False,
+        autoplay=True,
     )
 
     with gr.Blocks(title="OpenAI - Text to Speech") as demo:
@@ -125,24 +174,24 @@ def main():
                 )
 
             with gr.Column(scale=2):
+                # Initialize the input textbox with the desired label
                 input_textbox = gr.Textbox(
-                    label="Input Text",
+                    label="Input Text (0000 / 4096 chars)",
                     lines=10,
                     placeholder="Type your text here...",
                 )
-                # Add a character counter below the input textbox
-                char_count_text = gr.Markdown("0 / 4096")
 
-                # Function to update the character count
-                def update_char_count(input_text):
+                # Function to update the label with the character count
+                def update_label(input_text):
                     char_count = len(input_text)
-                    return f"**{char_count} / 4096**"
+                    new_label = f"Input Text ({char_count:04d} / 4096 chars)"
+                    return gr.update(label=new_label)
 
-                # Update character count when the user stops typing
+                # Update the label when the text changes
                 input_textbox.change(
-                    fn=update_char_count,
+                    fn=update_label,
                     inputs=input_textbox,
-                    outputs=char_count_text,
+                    outputs=input_textbox,
                 )
 
                 submit_button = gr.Button(
@@ -178,6 +227,5 @@ def main():
     # Launch the Gradio app with error display enabled
     demo.launch(show_error=True)
 
-
 if __name__ == "__main__":
-    main()
+    main()