Create app.py

app.py

@@ -0,0 +1,213 @@
+# app.py
+import gradio as gr
+import requests
+import random
+import urllib.parse
+import tempfile
+import os
+
+# --- Constants ---
+VOICES = [
+    "alloy", "echo", "fable", "onyx", "nova", "shimmer",  # Standard OpenAI Voices
+    "coral", "verse", "ballad", "ash", "sage", "amuch", "dan" # Additional Pollinations Voices? (Assuming based on list)
+]
+
+NSFW_URL_TEMPLATE = "https://text.pollinations.ai/Is this an inappropriate text-to-speech prompt \"{prompt}\". If yes then write \"YES\" only otherwise \"NO\" only"
+TTS_URL_TEMPLATE = "https://text.pollinations.ai/only repeat what i say now say with proper emphasis in a \"{emotion}\" emotion this statement - \"{prompt}\"?model=openai-audio&voice={voice}&seed={seed}"
+
+# --- Helper Functions ---
+
+def check_nsfw(prompt: str) -> bool:
+    """Checks if the prompt is NSFW using the Pollinations API."""
+    try:
+        # URL encode the prompt for safety
+        encoded_prompt = urllib.parse.quote(prompt)
+        url = NSFW_URL_TEMPLATE.format(prompt=encoded_prompt)
+        print(f"DEBUG: Checking NSFW URL: {url}") # Optional: for debugging
+
+        response = requests.get(url, timeout=20) # Added timeout
+        response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)
+
+        result = response.text.strip().upper()
+        print(f"DEBUG: NSFW Check Response: '{result}'") # Optional: for debugging
+
+        if result == "YES":
+            return True
+        elif result == "NO":
+            return False
+        else:
+            # Handle unexpected responses from the NSFW checker
+            print(f"Warning: Unexpected response from NSFW checker: {response.text}")
+            # Defaulting to safe might be risky, maybe default to NSFW? Or raise error?
+            # Let's default to considering it potentially unsafe if unsure.
+            return True # Treat unexpected responses as potentially NSFW
+
+    except requests.exceptions.RequestException as e:
+        print(f"Error during NSFW check: {e}")
+        # If the check fails, maybe treat as unsafe to be cautious
+        raise gr.Error(f"Failed to check prompt safety: {e}")
+    except Exception as e:
+        print(f"Unexpected error during NSFW check: {e}")
+        raise gr.Error(f"An unexpected error occurred during safety check: {e}")
+
+
+def generate_audio(prompt: str, voice: str, emotion: str, seed: int) -> bytes:
+    """Generates audio using the Pollinations Text-to-Speech API."""
+    try:
+        # URL encode the prompt and emotion
+        encoded_prompt = urllib.parse.quote(prompt)
+        encoded_emotion = urllib.parse.quote(emotion)
+
+        url = TTS_URL_TEMPLATE.format(
+            prompt=encoded_prompt,
+            emotion=encoded_emotion,
+            voice=voice,
+            seed=seed
+        )
+        print(f"DEBUG: Generating Audio URL: {url}") # Optional: for debugging
+
+        response = requests.get(url, timeout=60) # Increased timeout for audio generation
+        response.raise_for_status() # Raise an exception for bad status codes
+
+        # Check if response content type suggests audio
+        content_type = response.headers.get('content-type', '').lower()
+        if 'audio' not in content_type:
+            print(f"Warning: Unexpected content type received: {content_type}")
+            print(f"Response Text: {response.text[:500]}") # Log beginning of text response
+            raise gr.Error(f"API did not return audio. Response: {response.text[:200]}")
+
+        return response.content # Return raw audio bytes
+
+    except requests.exceptions.RequestException as e:
+        print(f"Error during audio generation: {e}")
+        # Try to get more info from response if available
+        error_details = ""
+        if hasattr(e, 'response') and e.response is not None:
+            error_details = e.response.text[:200] # Get first 200 chars of error response
+        raise gr.Error(f"Failed to generate audio: {e}. Details: {error_details}")
+    except Exception as e:
+        print(f"Unexpected error during audio generation: {e}")
+        raise gr.Error(f"An unexpected error occurred during audio generation: {e}")
+
+# --- Main Gradio Function ---
+
+def text_to_speech_app(prompt: str, voice: str, emotion: str, use_random_seed: bool, specific_seed: int):
+    """
+    Main function for the Gradio app. Checks NSFW, then generates audio.
+    Returns the path to a temporary audio file or an error message.
+    """
+    if not prompt:
+        raise gr.Error("Prompt cannot be empty.")
+    if not emotion:
+        # Default emotion if none provided, or raise error? Let's default.
+        emotion = "neutral"
+        print("Warning: No emotion provided, defaulting to 'neutral'.")
+        # raise gr.Error("Emotion cannot be empty.") # Alternative: require emotion
+    if not voice:
+         raise gr.Error("Please select a voice.")
+
+    # 1. Determine Seed
+    seed = random.randint(0, 2**32 - 1) if use_random_seed else int(specific_seed)
+    print(f"Using Seed: {seed}")
+
+    # 2. Check NSFW
+    print("Checking prompt safety...")
+    try:
+        is_nsfw = check_nsfw(prompt)
+    except gr.Error as e:
+        # Propagate errors raised by check_nsfw
+        return None, str(e) # Return None for audio, error message for text
+
+    if is_nsfw:
+        print("Prompt flagged as inappropriate.")
+        # Return None for audio output, and a message for a text output
+        return None, "Error: The prompt was flagged as inappropriate and cannot be processed."
+
+    # 3. Generate Audio (only if not NSFW)
+    print("Prompt is safe. Generating audio...")
+    try:
+        audio_bytes = generate_audio(prompt, voice, emotion, seed)
+
+        # 4. Save audio to a temporary file for Gradio
+        # Suffix is important for Gradio to recognize the format. Assuming MP3 based on common web usage.
+        # If the API returns WAV, change suffix to ".wav"
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_audio_file:
+            temp_audio_file.write(audio_bytes)
+            temp_file_path = temp_audio_file.name
+            print(f"Audio saved temporarily to: {temp_file_path}")
+
+        # Return the path to the temp file for the Audio component, and success message for Text
+        return temp_file_path, f"Audio generated successfully with voice '{voice}', emotion '{emotion}', and seed {seed}."
+
+    except gr.Error as e:
+        # Handle errors raised by generate_audio
+         return None, str(e) # Return None for audio, error message for text
+    except Exception as e:
+        print(f"Unexpected error in main function: {e}")
+        return None, f"An unexpected error occurred: {e}"
+
+
+# --- Gradio Interface ---
+
+def toggle_seed_input(use_random_seed):
+    """Updates the visibility of the specific seed input field."""
+    return gr.update(visible=not use_random_seed, value=12345) # Reset to default when shown
+
+with gr.Blocks() as app:
+    gr.Markdown("# Text-to-Speech with NSFW Check")
+    gr.Markdown(
+        "Enter text, choose a voice and emotion, and generate audio. "
+        "The text will be checked for appropriateness before generation."
+    )
+
+    with gr.Row():
+        with gr.Column(scale=2):
+            prompt_input = gr.Textbox(label="Prompt", placeholder="Enter the text you want to convert to speech...")
+            emotion_input = gr.Textbox(label="Emotion Style", placeholder="e.g., happy, sad, excited, calm...")
+            voice_dropdown = gr.Dropdown(label="Voice", choices=VOICES, value="alloy") # Default voice
+        with gr.Column(scale=1):
+            random_seed_checkbox = gr.Checkbox(label="Use Random Seed", value=True)
+            seed_input = gr.Number(label="Specific Seed", value=12345, visible=False, precision=0) # Integer seed
+
+    submit_button = gr.Button("Generate Audio", variant="primary")
+
+    with gr.Row():
+        audio_output = gr.Audio(label="Generated Audio", type="filepath") # Use filepath as we save temp file
+        status_output = gr.Textbox(label="Status") # To display errors or success messages
+
+    # --- Event Listeners ---
+    random_seed_checkbox.change(
+        fn=toggle_seed_input,
+        inputs=[random_seed_checkbox],
+        outputs=[seed_input]
+    )
+
+    submit_button.click(
+        fn=text_to_speech_app,
+        inputs=[
+            prompt_input,
+            voice_dropdown,
+            emotion_input,
+            random_seed_checkbox,
+            seed_input
+        ],
+        outputs=[audio_output, status_output] # Output to both components
+    )
+
+    gr.Examples(
+        examples=[
+            ["Hello there! This is a test of the text-to-speech system.", "alloy", "neutral", True, 12345],
+            ["What a beautiful day to build Gradio apps.", "shimmer", "happy", True, 12345],
+            ["I am feeling a bit down today.", "fable", "sad", False, 9876],
+            ["This technology is absolutely amazing!", "nova", "excited", True, 12345],
+        ],
+        inputs=[prompt_input, voice_dropdown, emotion_input, random_seed_checkbox, seed_input],
+        outputs=[audio_output, status_output], # Outputs match the click function
+        fn=text_to_speech_app, # The function to call for examples
+        cache_examples=False, # Might be good to disable caching if APIs change or have quotas
+    )
+
+# --- Launch the App ---
+if __name__ == "__main__":
+    app.launch()
+