import gradio as gr import os import io import requests, json from PIL import Image import base64 from dotenv import load_dotenv, find_dotenv _ = load_dotenv(find_dotenv()) # read local .env file hf_api_key = os.environ['HF_API_KEY'] # Load the translation model (Turkish to English) API_URL = "https://api-inference.huggingface.co/models/Helsinki-NLP/opus-mt-tr-en" headers = { "Authorization": f"Bearer {hf_api_key}", "Content-Type": "application/json" } # Text-to-image endpoint def get_completion(inputs, parameters=None, ENDPOINT_URL=os.environ['HF_API_TTI_STABILITY_AI']): data = {"inputs": inputs} if parameters is not None: data.update({"parameters": parameters}) response = requests.post(ENDPOINT_URL, headers=headers, data=json.dumps(data)) # Check the content type of the response content_type = response.headers.get('Content-Type', '') print(content_type) if 'application/json' in content_type: return json.loads(response.content.decode("utf-8")) elif 'image/' in content_type: return response.content # return raw image data response.raise_for_status() # raise an error for unexpected content types # A helper function to convert the PIL image to base64 def base64_to_pil(img_base64): base64_decoded = base64.b64decode(img_base64) byte_stream = io.BytesIO(base64_decoded) pil_image = Image.open(byte_stream) return pil_image def query(payload): response = requests.post(API_URL, headers=headers, json=payload) return response.json() # Translation function def translate_to_english(text): try: # Translate the input from Turkish to English translation = query({"inputs": text}) print(translation) translated_text = translation[0]['translation_text'] return translated_text except Exception as e: print(f"Translation error: {e}") return text # If translation fails, return original text # Main generation function with translation def generate(prompt, negative_prompt, steps, guidance, width, height): # Translate the prompt to English if it's in Turkish translated_prompt = translate_to_english(prompt) print(f"Translated Prompt: {translated_prompt}") params = { "negative_prompt": negative_prompt, "num_inference_steps": steps, "guidance_scale": guidance, "width": width, "height": height } output = get_completion(translated_prompt, params) # Check if the output is an image (bytes) or JSON (dict) if isinstance(output, dict): raise ValueError("Expected an image but received JSON: {}".format(output)) # If output is raw image data, convert it to a PIL image result_image = Image.open(io.BytesIO(output)) return (translated_prompt, result_image) with gr.Blocks() as demo: gr.Markdown("## Image Generation with Turkish Inputs") gr.Markdown("### [`stabilityai/stable-diffusion-xl-base-1.0`](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) and [`Helsinki-NLP/opus-mt-tr-en`](https://huggingface.co/Helsinki-NLP/opus-mt-tr-en) models work under the hood!") with gr.Row(): with gr.Column(scale=4): prompt = gr.Textbox(label="Your prompt (in Turkish or English)") # Accept Turkish or English input with gr.Column(scale=1, min_width=50): btn = gr.Button("Submit") with gr.Accordion("Advanced options", open=False): negative_prompt = gr.Textbox(label="Negative prompt") with gr.Row(): with gr.Column(): steps = gr.Slider(label="Inference Steps", minimum=1, maximum=100, value=50, info="In how many steps will the denoiser denoise the image?") guidance = gr.Slider(label="Guidance Scale", minimum=1, maximum=20, value=7, info="Controls how much the text prompt influences the result") with gr.Column(): width = gr.Slider(label="Width", minimum=64, maximum=1024, step=64, value=512) height = gr.Slider(label="Height", minimum=64, maximum=1024, step=64, value=512) translated_text = gr.Textbox(label="Translated text") output = gr.Image(label="Result") btn.click(fn=generate, inputs=[prompt, negative_prompt, steps, guidance, width, height], outputs=[translated_text, output]) gr.close_all() demo.launch(share=True)