app.py

import gradio as gr
import torch
import numpy as np
from PIL import Image
from transformers import (
    pipeline,
    AutoModelForCausalLM,
    AutoProcessor,
    AutoTokenizer,
    GenerationConfig,
    TextStreamer,
)
from datasets import load_dataset

# Use CPU if no GPU is available
device = "cuda" if torch.cuda.is_available() else "cpu"
dtype = torch.float16 if torch.cuda.is_available() else torch.float32
print(f"Device set to use {device}")

# Load image captioning model (BLIP)
caption_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base", device=device)

# Load OCR model (Florence-2-base)
ocr_model = AutoModelForCausalLM.from_pretrained(
    "microsoft/Florence-2-base", trust_remote_code=True, torch_dtype=dtype
).to(device)
ocr_processor = AutoProcessor.from_pretrained("microsoft/Florence-2-base", trust_remote_code=True)

# Load SmallDoge model for context generation
doge_tokenizer = AutoTokenizer.from_pretrained("SmallDoge/Doge-320M-Instruct")
doge_model = AutoModelForCausalLM.from_pretrained(
    "SmallDoge/Doge-320M-Instruct", trust_remote_code=True
).to(device)
doge_config = GenerationConfig(
    max_new_tokens=100,
    use_cache=True,
    do_sample=True,
    temperature=0.8,
    top_p=0.9,
    repetition_penalty=1.0
)

# Load SpeechT5 for TTS
synthesiser = pipeline("text-to-speech", model="microsoft/speecht5_tts", device=device)

# Use known compatible 600-dim speaker embedding
embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
speaker_embedding = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)  # Shape: [1, 600]

def process_image(image):
    try:
        # Caption generation
        caption = caption_model(image)[0]['generated_text']

        # OCR extraction
        ocr_inputs = ocr_processor(text="<OCR>", images=image, return_tensors="pt").to(device)
        ocr_outputs = ocr_model.generate(
            input_ids=ocr_inputs["input_ids"],
            pixel_values=ocr_inputs["pixel_values"],
            max_new_tokens=1024,
            num_beams=3,
            do_sample=False,
        )
        extracted_text = ocr_processor.batch_decode(ocr_outputs, skip_special_tokens=True)[0]

        # Context generation using Doge
        prompt = f"Determine the context of this image based on the caption and extracted text.\nCaption: {caption}\nExtracted text: {extracted_text}\nContext:"
        conversation = [{"role": "user", "content": prompt}]
        doge_inputs = doge_tokenizer.apply_chat_template(conversation, tokenize=True, return_tensors="pt").to(device)
        doge_output = doge_model.generate(doge_inputs, generation_config=doge_config)
        context = doge_tokenizer.decode(doge_output[0], skip_special_tokens=True)

        # Convert context to speech
        speech = synthesiser(
            context,
            forward_params={"speaker_embeddings": speaker_embedding}
        )
        audio = np.array(speech["audio"])
        rate = speech["sampling_rate"]

        return (rate, audio), caption, extracted_text, context

    except Exception as e:
        return None, f"Error: {str(e)}", "", ""

# Gradio UI
iface = gr.Interface(
    fn=process_image,
    inputs=gr.Image(type='pil', label="Upload an Image"),
    outputs=[
        gr.Audio(label="Generated Audio"),
        gr.Textbox(label="Generated Caption"),
        gr.Textbox(label="Extracted Text (OCR)"),
        gr.Textbox(label="Generated Context")
    ],
    title="SeeSay Contextualizer with Doge + SpeechT5",
    description="Upload an image to generate a caption, extract OCR text, determine context with Doge-320M, and hear it with SpeechT5."
)

iface.launch(share=True)