Create app.py

app.py

@@ -0,0 +1,35 @@
+import gradio as gr
+from transformers import pipeline
+from PIL import Image
+import numpy as np
+from datasets import load_dataset
+import soundfile as sf
+import torch
+
+image_to_text = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")
+synthesiser = pipeline("text-to-speech", "microsoft/speecht5_tts")
+
+embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
+speaker_embedding = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
+
+def predict_step(image):
+    if isinstance(image, np.ndarray):
+        image = Image.fromarray(image)
+    
+  
+    result = image_to_text(image)
+    
+    texto = result[0]['generated_text'] 
+    speech = synthesiser(texto, forward_params={"speaker_embeddings": speaker_embedding})
+    sf.write("speech.wav", speech["audio"], samplerate=speech["sampling_rate"])
+    return "speech.wav", texto
+
+demo = gr.Interface(
+    fn=predict_step, 
+    inputs="image", 
+    outputs=["audio","textbox"],
+    title="Descripción de Imágenes",
+    description="Cargue una imagen y obtenga una descripción generada por IA."
+)
+
+demo.launch()