Update app.py

app.py

@@ -14,134 +14,109 @@ logger = logging.getLogger(__name__)
 MAX_TEXT_LENGTH = 5000
 TEMP_FILES = []
 
-# Инициализация клиентов (будет выполнена после получения API ключа)
+# Инициализация клиентов
 image_client = None
 chat_client = None
+tts_voices = []
 
 # Стиль интерфейса
 theme = gr.themes.Soft()
 
 def initialize_clients(api_key):
-    """Инициализирует клиенты API с полученным ключом"""
-    global image_client, chat_client
+    global image_client, chat_client, tts_voices
     try:
+        # Инициализация клиентов API
         image_client = InferenceClient(token=api_key)
         chat_client = InferenceClient(provider="together", api_key=api_key)
+        
+        # Получение списка голосов для TTS
+        asyncio.run(load_tts_voices())
         return True
     except Exception as e:
         logger.error(f"Initialization error: {str(e)}")
         return False
 
-# Парсер промптов
-def parse_image_prompts(text):
-    return re.findall(r'<image>(.*?)<\/image>', text, re.DOTALL)
+async def load_tts_voices():
+    global tts_voices
+    voices = await edge_tts.list_voices()
+    tts_voices = [f"{v['ShortName']} ({v['Gender']})" for v in voices]
+
+# Функция очистки текста
+def clean_text(text):
+    text = re.sub(r'[*_~><`#%$^]', '', text)
+    return text.strip()
 
-# Генератор изображений
-def generate_image(prompt, api_key):
-    """Генерирует изображение с проверкой API ключа"""
+# Генерация аудио
+async def generate_audio(text, voice, api_key):
     if not api_key:
-        return None, "⚠️ Please enter your API key first"
+        return None, "⚠️ Please enter API key first"
     
-    if not initialize_clients(api_key):
-        return None, "⚠️ Invalid API key"
+    if not text.strip():
+        return None, "⚠️ Please enter text"
     
     try:
-        response = image_client.text_to_image(
-            prompt.strip(),
-            model="nerijs/dark-fantasy-illustration-flux",
-            negative_prompt="text, watermark, low quality",
-            guidance_scale=9,
-            height=512,
-            width=512
-        )
+        text = clean_text(text)
+        communicate = edge_tts.Communicate(text, voice.split()[0])
         
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".png") as tmp_file:
-            tmp_file.write(response)
-            TEMP_FILES.append(tmp_file.name)
-            return tmp_file.name, None
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
+            tmp_path = tmp_file.name
+            await communicate.save(tmp_path)
+            TEMP_FILES.append(tmp_path)
+            return tmp_path, None
             
     except Exception as e:
-        logger.error(f"Image error: {str(e)}")
-        return None, f"⚠️ Error: {str(e)}"
+        logger.error(f"TTS error: {str(e)}")
+        return None, f"⚠️ Audio error: {str(e)}"
 
-# Генератор текста
-async def generate_text(input_text, api_key):
-    """Генерирует текст с проверкой API ключа"""
-    if not api_key:
-        return None, "⚠️ Please enter your API key first"
-    
-    if not initialize_clients(api_key):
-        return None, "⚠️ Invalid API key"
-    
-    try:
-        completion = chat_client.chat.completions.create(
-            model="deepseek-ai/DeepSeek-R1",
-            messages=[{"role": "user", "content": input_text}],
-            max_tokens=500
-        )
-        return completion.choices[0].message.content, None
-    except Exception as e:
-        logger.error(f"Text error: {str(e)}")
-        return None, f"⚠️ Error: {str(e)}"
+# Остальные функции (generate_image, generate_text) остаются без изменений...
 
-# Интерфейс
 def create_interface():
-    with gr.Blocks(theme=theme, title="AI Studio") as demo:
-        gr.Markdown("# 🔑 AI Content Generator")
+    with gr.Blocks(theme=theme, title="AI Studio Pro") as demo:
+        gr.Markdown("# 🎙️ Full AI Studio")
         
-        # Секция для ввода API ключа
-        with gr.Row():
-            api_key = gr.Textbox(
-                label="HuggingFace API Key",
-                type="password",
-                placeholder="Enter your HF_API_KEY here...",
-                info="Get your key from https://huggingface.co/settings/tokens"
-            )
+        # Секция API ключа
+        api_key = gr.Textbox(
+            label="HuggingFace API Key",
+            type="password",
+            placeholder="Enter HF_API_KEY...",
+            info="Get key from https://huggingface.co/settings/tokens"
+        )
         
-        # Основная секция
-        with gr.Row():
-            with gr.Column():
-                input_text = gr.Textbox(
-                    label="Input Text",
-                    placeholder="Enter text with <image>prompt</image> tags...",
-                    lines=5
-                )
-                
+        with gr.Tabs():
+            # Вкладка TTS
+            with gr.Tab("🔊 Text-to-Speech"):
                 with gr.Row():
-                    text_btn = gr.Button("Generate Text", variant="primary")
-                    image_btn = gr.Button("Generate Images", variant="secondary")
+                    with gr.Column():
+                        tts_text = gr.Textbox(label="Text to speak", lines=5)
+                        tts_voice = gr.Dropdown(
+                            label="Voice",
+                            choices=tts_voices,
+                            interactive=True
+                        )
+                        tts_btn = gr.Button("Generate Speech", variant="primary")
+                    
+                    with gr.Column():
+                        tts_audio = gr.Audio(label="Generated Audio")
+                        tts_status = gr.HTML()
                 
-            with gr.Column():
-                output_text = gr.Textbox(label="Generated Text", interactive=False)
-                gallery = gr.Gallery(label="Generated Images", columns=2)
-                status = gr.HTML()
+                tts_btn.click(
+                    generate_audio,
+                    [tts_text, tts_voice, api_key],
+                    [tts_audio, tts_status]
+                )
 
-        # Обработчики событий
-        text_btn.click(
-            generate_text,
-            inputs=[input_text, api_key],
-            outputs=[output_text, status]
-        )
-        
-        image_btn.click(
-            lambda text, key: [generate_image(p, key) for p in parse_image_prompts(text)],
-            inputs=[output_text, api_key],
-            outputs=[gallery, status]
-        )
+            # Вкладка генерации текста и изображений
+            with gr.Tab("🖼️ Text & Image Generation"):
+                # ... предыдущий код вкладки ...
 
-    return demo
+        return demo
 
 # Запуск
 if __name__ == "__main__":
     try:
         demo = create_interface()
-        demo.launch(
-            server_port=7860,
-            show_error=True,
-            share=False
-        )
+        demo.launch(server_port=7860)
     finally:
-        # Очистка временных файлов
         for file in TEMP_FILES:
             try:
                 Path(file).unlink(missing_ok=True)