1

app.py

@@ -6,20 +6,21 @@ def text2speech(text):
     return fastspeech(text)
     
 def engine(text_input):
-    #ner = gr.Interface.load("huggingface/flair/ner-english-ontonotes-large")
-    #entities = ner(text_input)
-    #entities = [tupl for tupl in entities if None not in tupl]
-    #entities_num = len(entities)
+    ner = gr.Interface.load("huggingface/flair/ner-english-ontonotes-large")
+    entities = ner(text_input)
+    entities = [tupl for tupl in entities if None not in tupl]
+    entities_num = len(entities)
     
     #img = run(text_input,'50','256','256','1',10)  #entities[0][0]
+    
     img = gr.Interface.load("spaces/multimodalart/latentdiffusion")(text_input,'50','256','256','1',10)[0] #inputs={'prompt':text_input,'steps':'50','width':'256','height':'256','images':'1','scale':10}).launch()
     #img_intfc = gr.Interface.load("spaces/multimodalart/latentdiffusion", inputs=[gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text"), gr.inputs.Textbox(lines=1, label="Input Text")],
     #outputs=[gr.outputs.Image(type="pil", label="output image"),gr.outputs.Carousel(label="Individual images",components=["image"]),gr.outputs.Textbox(label="Error")], )
     #title="Convert text to image")
     #img = img_intfc[0]
     #img = img_intfc(text_input,'50','256','256','1',10)
-    print(img)
-    print(type(img))
+    #print(img)
+    #print(type(img))
     #print(img)
     #print(type(img[1][0][0]))
     #print(img[1])
@@ -27,13 +28,13 @@ def engine(text_input):
     #inputs=['George',50,256,256,1,10]
     #run(prompt, steps, width, height, images, scale)
     
-    #speech = text2speech(text_input)
-    return img #entities, speech, img
+    speech = text2speech(text_input)
+    return img, entities, speech
     
 #image = gr.outputs.Image(type="pil", label="output image")
 app = gr.Interface(engine, 
                    gr.inputs.Textbox(lines=5, label="Input Text"),
-                   gr.outputs.Image(type="auto", label="Output"),
+                   [gr.outputs.Image(type="auto", label="Output"), gr.outputs.Textbox(type="auto", label="Text"), gr.outputs.Audio(type="file", label="Speech Answer") ],
                    #live=True,
                    #outputs=[#gr.outputs.Textbox(type="auto", label="Text"),gr.outputs.Audio(type="file", label="Speech Answer"),
                    #outputs= img, #gr.outputs.Carousel(label="Individual images",components=["image"]), #, gr.outputs.Textbox(label="Error")],