Update app.py

app.py

@@ -3,159 +3,147 @@ import edge_tts
 import asyncio
 import tempfile
 import re
-from typing import Optional, Tuple, Dict
+import emoji
+import requests
+from PIL import Image
+from io import BytesIO
 from huggingface_hub import InferenceClient
+import os
+from dotenv import load_dotenv
 
-# Константы
-DEFAULT_RATE = 0
-DEFAULT_PITCH = 0
-HF_API_KEY = "YOUR_API_KEY"  # Замените на ваш API ключ
+# Загрузка переменных окружения из .env файла
+load_dotenv()
 
-# Инициализация клиента для генерации текста
-text_client = InferenceClient(provider="together", api_key=HF_API_KEY)
+# Функция для очистки текста от нежелательных символов и эмодзи
+def clean_text(text):
+    text = re.sub(r'[*_~><]', '', text)
+    text = emoji.replace_emoji(text, replace='')
+    return text
 
-# Кастомная цветовая схема
-theme = gr.themes.Default(
-    primary_hue="orange",
-    secondary_hue="yellow",
-).set(
-    button_primary_background="linear-gradient(90deg, #ff9a00, #ffd700)",
-    button_primary_background_hover="linear-gradient(90deg, #ff8c00, #ffcc00)",
-    slider_color="#ff9a00",
-    block_background="#fff5e6"
-)
-
-async def get_voices() -> Dict[str, str]:
+# Get all available voices
+async def get_voices():
     voices = await edge_tts.list_voices()
-    return {
-        f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName']
-        for v in voices
-    }
+    return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
 
-async def text_to_speech(text: str, voice: str, rate: int, pitch: int) -> Tuple[Optional[str], Optional[str]]:
+# Text-to-speech function
+async def text_to_speech(text, voice, rate, pitch):
     if not text.strip():
-        return None, None
-    try:
-        voice_short_name = voice.split(" - ")[0]
-        communicate = edge_tts.Communicate(
-            text, 
-            voice_short_name,
-            rate=f"{rate:+d}%",
-            pitch=f"{pitch:+d}Hz"
-        )
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
-            await communicate.save(tmp_file.name)
-            return tmp_file.name, None
-    except Exception:
-        return None, None
-
-# ADD PROMPT HERE - Вставьте ваш промпт для генерации текста
-SYSTEM_PROMPT = """
-[Здесь будет ваш системный промпт]
-"""
+        return None, "Please enter text to convert."
+    if not voice:
+        return None, "Please select a voice."
+    
+    text = clean_text(text)
+    voice_short_name = voice.split(" - ")[0]
+    rate_str = f"{rate:+d}%"
+    pitch_str = f"{pitch:+d}Hz"
+    communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
+        tmp_path = tmp_file.name
+        try:
+            await communicate.save(tmp_path)
+        except Exception as e:
+            return None, f"An error occurred during text-to-speech conversion: {str(e)}"
+    return tmp_path, None
 
-async def generate_detailed_description(input_text: str) -> str:
-    try:
-        # Добавьте вашу логику обработки промпта
-        messages = [
-            {"role": "system", "content": SYSTEM_PROMPT},
-            {"role": "user", "content": input_text}
-        ]
-        
-        completion = text_client.chat.completions.create(
-            model="deepseek-ai/DeepSeek-R1",
-            messages=messages,
-            max_tokens=500
-        )
-        return completion.choices[0].message.content
-    except Exception:
-        return ""
+# Generate image using your custom prompt and API
+def generate_image_with_prompt(prompt, api_key):
+    client = InferenceClient(
+        provider="together",
+        api_key=api_key
+    )
+    messages = [
+        {
+            "role": "user",
+            "content": prompt
+        }
+    ]
+    completion = client.chat.completions.create(
+        model="deepseek-ai/DeepSeek-R1", 
+        messages=messages, 
+        max_tokens=500
+    )
+    response = completion.choices[0].message.content
+    # Assuming the response contains a URL or image data
+    if "http" in response:
+        image_url = response
+        image_response = requests.get(image_url)
+        image = Image.open(BytesIO(image_response.content))
+        return image
+    else:
+        return None, "Failed to generate image."
 
-async def tts_interface(text: str, voice: str, rate: int, pitch: int):
-    return await text_to_speech(text, voice, rate, pitch)
+# Gradio interface function
+def tts_interface(*args):
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+    audio, warning = loop.run_until_complete(text_to_speech(*args[:-2]))
+    prompt = args[-2]
+    image = generate_image_with_prompt(prompt, args[-1])
+    loop.close()
+    if warning:
+        return (None, gr.update(value=f"<span style='color:red;'>{warning}</span>", visible=True), None)
+    return (audio, gr.update(visible=False), image)
 
-async def create_demo():
+# Create Gradio application
+async def create_demo(api_key):
     voices = await get_voices()
     
-    with gr.Blocks(theme=theme, title="AI Studio Pro") as demo:
-        gr.Markdown("# 🧡💛 AI Creative Studio")
-        
-        # Вкладки
-        with gr.Tabs():
-            # Вкладка TTS
-            with gr.Tab("🔊 Генерация речи"):
-                with gr.Row():
-                    with gr.Column():
-                        text_input = gr.Textbox(
-                            label="Входной текст",
-                            lines=5,
-                            placeholder="Введите текст для озвучки...",
-                            elem_classes="orange-border"
-                        )
-                        with gr.Row():
-                            lang_dropdown = gr.Dropdown(
-                                choices=["Все языки", "en", "ru", "es", "fr", "de"],
-                                value="Все языки",
-                                label="Выберите язык"
-                            )
-                            voice_dropdown = gr.Dropdown(
-                                choices=list(voices.keys()),
-                                label="Выберите голос",
-                                interactive=True
-                            )
-                        with gr.Row():
-                            rate_slider = gr.Slider(-50, 50, 0, label="Скорость")
-                            pitch_slider = gr.Slider(-20, 20, 0, label="Тон")
-                        generate_btn = gr.Button("Сгенерировать речь", variant="primary")
-                    
-                    with gr.Column():
-                        audio_output = gr.Audio(label="Результат", elem_classes="orange-border")
-        
-            # Вкладка генерации контента
-            with gr.Tab("✨ Генератор контента"):
-                with gr.Row():
-                    with gr.Column():
-                        prompt_input = gr.Textbox(
-                            label="Краткое описание",
-                            lines=3,
-                            placeholder="Опишите идею для генерации...",
-                            elem_classes="orange-border"
-                        )
-                        with gr.Accordion("Дополнительные настройки", open=False):
-                            gr.Markdown("Здесь будут дополнительные параметры")
-                        generate_content_btn = gr.Button("Сгенерировать контент", variant="primary")
-                    
-                    with gr.Column():
-                        text_output = gr.Textbox(
-                            label="Сгенерированный текст",
-                            interactive=False,
-                            elem_classes="orange-border"
-                        )
-                        image_output = gr.Gallery(
-                            label="Сгенерированные изображения",
-                            columns=2,
-                            elem_classes="orange-border"
-                        )
+    description = """
+    This is a simple text-to-speech and image generation application.
+    Enter any text to convert it into speech and describe an image to generate it.
+    Adjust the speech rate and pitch according to your preference.
+    """
 
-        # Обработчики событий
-        generate_btn.click(
-            tts_interface,
-            [text_input, voice_dropdown, rate_slider, pitch_slider],
-            audio_output
-        )
-        
-        generate_content_btn.click(
-            generate_detailed_description,
-            prompt_input,
-            text_output
-        )
+    css = """
+    .gradio-container {
+        background: linear-gradient(135deg, #e0eafc, #cfdef3);
+        color: #333;
+    }
+    .gr-button {
+        background-color: #007aff;
+        color: white;
+    }
+    .gr-button:hover {
+        background-color: #005bb5;
+    }
+    .gr-textbox, .gr-slider, .gr-dropdown {
+        border-radius: 8px;
+        padding: 10px;
+        margin: 10px 0;
+    }
+    .gr-markdown {
+        color: red;
+    }
+    """
 
+    demo = gr.Interface(
+        fn=tts_interface,
+        inputs=[
+            gr.Textbox(label="Input Text", lines=5),
+            gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value=""),
+            gr.Slider(minimum=-50, maximum=50, value=0, label="Speech Rate Adjustment (%)", step=1),
+            gr.Slider(minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1),
+            gr.Textbox(label="Image Prompt", lines=2),
+            gr.Textbox(label="Hugging Face API Key", value=api_key, type="password")
+        ],
+        outputs=[
+            gr.Audio(label="Generated Audio", type="filepath"),
+            gr.Markdown(label="Warning", visible=False),
+            gr.Image(label="Generated Image")
+        ],
+        title="Edge TTS Text-to-Speech & Image Generation",
+        description=description,
+        css=css,
+        article="",
+        analytics_enabled=False,
+        allow_flagging="manual"
+    )
     return demo
 
-async def main():
-    demo = await create_demo()
-    demo.queue()
-    demo.launch()
-
+# Run the application
 if __name__ == "__main__":
-    asyncio.run(main())
+    api_key = os.getenv('HUGGING_FACE_API_KEY')  # Ваш API ключ
+    if not api_key:
+        raise ValueError("Please set the HUGGING_FACE_API_KEY environment variable.")
+    demo = asyncio.run(create_demo(api_key))
+    demo.launch()