Update app.py

app.py

@@ -111,14 +111,14 @@ def inference(
     task,
     text_list=None,
     image=None,
-    audio=None,
     image2=None,
+    audio=None,
 ):
     if task == "image-text":
         result = image_text_zeroshot(image, text_list)
     elif task == "audio-text":
         result = audio_text_zeroshot(audio, text_list)
-    elif task == "video-text":
+    elif task == "embeddings of two images and textlist":
         result = doubleimage_text_zeroshot(image, image2, text_list)
     else:
         raise NotImplementedError
@@ -131,7 +131,7 @@ def main():
             choices=[
                 "image-text",
                 "audio-text",
-                "video-text",
+                "embeddings of two images and textlist",
             ],
             type="value",
             default="image-text",
@@ -139,18 +139,16 @@ def main():
         ),
         gr.inputs.Textbox(lines=1, label="Candidate texts"),
         gr.inputs.Image(type="filepath", label="Input image"),
-        gr.inputs.Audio(type="filepath", label="Input audio"),
         gr.inputs.Image(type="filepath", label="Input image2"),
+        gr.inputs.Audio(type="filepath", label="Input audio"),
+    
     ]
 
     iface = gr.Interface(
         inference,
         inputs,
         "label",
-        description="""<p>This is a simple demo of ImageBind for zero-shot cross-modal understanding (now including image classification, audio classification, and video classification). Please refer to the original <a href='https://arxiv.org/abs/2305.05665' target='_blank'>paper</a> and <a href='https://github.com/facebookresearch/ImageBind' target='_blank'>repo</a> for more details.<br>
-                    To test your own cases, you can upload an image, an audio or a video, and provide the candidate texts separated by "|".<br>
-                    You can duplicate this space and run it privately: <a href='https://huggingface.co/spaces/OFA-Sys/chinese-clip-zero-shot-image-classification?duplicate=true'><img src='https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14' alt='Duplicate Space'></a></p>""",
-        title="ImageBind: Zero-shot Cross-modal Understanding",
+        title="Multimodal AI assitive agents for Learning Disorders : Demo with embeddings of ImageBind: ",
     )
 
     iface.launch()